B7D9EG6C9
2017-10-06 05:14:18
[cucumber/cucumber] aslakhellesoy requested changes to #223: gherkin/java:Support unicode colon as separator.
<p>Please remove the old <code>TITLE_KEYWORD_SEPARATOR</code> and rename <code>TITLE_KEYWORD_SEPARATOR2</code> to <code>TITLE_KEYWORD_SEPARATOR</code></p>
<p>This change must be ported to all other Gherkin implementations (Go, C, Ruby, JavaScript etc) before it can be merged.</p>
B7D9EG6C9
2017-10-06 06:54:18
B7D9EG6C9
2017-10-11 08:03:20
[cucumber/cucumber] vcavallo requested changes to #50: Modified documents and deleted images
<p>> Behavior Driven
Behavior-Driven</p>
<p>> is an software development
is a software development</p>
<p>> Behavior driven
Behavior-driven</p>
<p>> Once the test passes, observe the resulting design and re-factor to remove any duplication you see.
suggestion:</p>
<p>Once the test passes, observe the resulting design and re-factor according to whatever software design principles seem to best apply.</p>
<hr />
<p><code>DRY</code> isn't the only thing to pay attention to ;)</p>
<p>> domain driven design
domain-driven</p>
<p>~Test Driven Development is a developer practice that involves writing tests before writing the code being tested. Begin by writing a very small test for code that does not yet exist. Run the test and, naturally, it fails. Now write just enough code to make that test pass. No more. Once the test passes, observe the~</p>
<p>remove this entirely. it's a repetition of above.</p>
B7D9EG6C9
2017-10-12 09:28:19
B7D9EG6C9
2017-10-21 02:28:15
[cucumber/cucumber] mpkorstanje commented on #291: [WIP] Add datatable/java
<p>I have changed the to <code>match(String text, List&lt;List&lt;String&gt;&gt; tableArgument)</code> so callers don't have to provide their own data table with converter which needs a parameter type registry instance. This simplifies the logistics of using Expressions.</p>
B7D9EG6C9
2017-10-21 02:35:17
[cucumber/cucumber] mpkorstanje commented on #291: [WIP] Add datatable/java
<p>We can extract the <code>Transformer</code> and leave its creation to the caller. If we also provide the default table transform we can cut the dependency between datatable and cucumber-expressions. </p>
B7D9EG6C9
2017-10-21 16:49:20
[cucumber/cucumber] mpkorstanje commented on #291: [wip] add datatable/java
<p>This is wrong. We should use the match method first and pass that result list to the value converter.</p>
B7D9EG6C9
2017-10-22 07:39:15
[cucumber/cucumber] brasmusson commented on #292: gherkin: Pass the content type of a docstring down into its pickle string form…
<p>This file should not be edited directly, it is <a href="https://github.com/cucumber/cucumber/blob/master/gherkin/testdata/good/docstrings.feature.pickles.ndjson">gherkin/testdata/good/docstrings.feature.pickles.ndjson</a> which is the source file that is pushed down to each language version of the gherkin library (using <a href="https://github.com/cucumber/cucumber/blob/master/scripts/functions.sh#L54-L63">rsync_files</a>). Since we use the same testdata for acceptance tests for all language version for the gherkin library, this pull-request cannot be merged until all 8 language versions have been updated.</p>
B7D9EG6C9
2017-10-25 08:39:20
[cucumber/cucumber] mlvandijk commented on #50: Modified documents and deleted images
<p>I'd like to suggest removing the last 2 paragraphs (from "Let me try to explain...") entirely. Imho the first paragraph is clear enough.</p>
<p>"procdure" should be "procedure"</p>
B7D9EG6C9
2017-12-06 12:21:20
[cucumber/cucumber] mpkorstanje commented on #313: gherkin java: Extended the languages list to include a column with the language name and a column with the native name
<p>I think that <code>assertThat("Swedish with explanation should be available", actual, CoreMatchers.containsString(expected))</code> is a bit more readable and provides better diagnostics when the string doesn't contain the expected value. </p>
<p>Unfortunately it looks like we're using <code>mockito-all</code> in this project which bundles a version of <code>hamcrest-core</code> that doesn't have <code>containsString</code>. </p>
<p>I'll add a commit to resolve this</p>
B7D9EG6C9
2017-12-07 01:15:25
[cucumber/cucumber] tsundberg commented on #313: gherkin java: Extended the languages list to include a column with the language name and a column with the native name
<p>Not really. I haven't seen it myself before either.</p>
<p>I expected that the result from <code>keywords.get("name");</code> would be a list. It turns out that it isn't. It's a String. But as it is declared as a <code>List&lt;String&gt;</code>, the compiler will not allow me to cast it to a <code>String</code>. My first attempt was divided on two lines and casts. Then I inlined the result and ended up here. I was puzzled for a while and then I gave up. I blaim the internals of GSON.</p>
<p>The attached screenshot is from when I looked at af with a debugger.</p>
<p><a href="https://user-images.githubusercontent.com/541840/33707257-24c75770-db37-11e7-990e-823e988cc911.png">screen shot 2017-12-07 at 10 10 09</a></p>
B7D9EG6C9
2018-01-19 14:35:23
B7D9EG6C9
2018-01-20 13:47:17
[cucumber/cucumber] mpkorstanje commented on #313: gherkin java: Extended the languages list to include a column with the language name and a column with the native name
<p>There reason this happens is because of an unchecked cast in GherkinDialectProvider which forces the type to be a <code>Map&lt;String, Map&lt;String, List&lt;String&gt;&gt;&gt;</code>. </p>
<p>However the data structure is not actually a <code>Map&lt;String, Map&lt;String, List&lt;String&gt;&gt;&gt;</code> but <code>Map&lt;String, Map&lt;String, Object&gt;&gt;</code> as the properties of a language in <code>gherking-languages.json</code> are either a list or a string.</p>
<p>This can be avoided by using the <code>fromJson</code> that accepts a type as input and using <code>Map&lt;String, Map&lt;String, Object&gt;&gt;</code> as type.</p>
B7D9EG6C9
2018-03-05 13:19:25
[cucumber/cucumber] aslakhellesoy commented on #341: update docker to support go builds
<p>This wasn't necessary for building <code>gherkin/go</code>. Is it really necessary for <code>tag-expressions/go</code>? I'd like to avoid too many language-isms in this Docker image. </p>
<p>It's supposed to be a kitchen sink with all language dependencies, but without favouring any particular one.</p>
B7D9EG6C9
2018-03-06 10:47:27
[cucumber/cucumber] charlierudolph commented on #339: tag expressions: add go implementation
<p>This is what was done in gherkin/go. This is never something I have done and think is too much against the go way of doing things. I would prefer we setup go to have its own docker image</p>
B7D9EG6C9
2018-03-06 12:03:29
B7D9EG6C9
2018-03-06 12:04:25
B7D9EG6C9
2018-04-08 15:10:25
[cucumber/cucumber] aslakhellesoy requested changes to #342: tag-expressions for python
<p>Delete this?</p>
<p>The build needs to run without <code>sudo</code> - it should be non-interactive and shouldn't make global modifications to the system.</p>
<p>The other python projects don't use <code>sudo</code> - can you follow the same convention?</p>
B7D9EG6C9
2018-04-19 13:57:25
[cucumber/cucumber] jenisys commented on #342: tag-expressions for python
<p>The makefile use no "sudo" on Travis, if this is what is meant what you meant.
The makefile checks all assumptions and needs to install a python module "pipenv" if it is not installed yet. I had initially a simpler solution but it was not working on Travis or otherwise not working in a normal user workspace.</p>
<p>NOTE: Look at the complete "Makefile". If you have any questions, we can talk about it.</p>
B7D9EG6C9
2018-04-19 14:01:26
[cucumber/cucumber] jenisys commented on #342: tag-expressions for python
<p>Removing the import-line will break the code and make all tag-parser related tests fail. These symbols are needed in the parser module.
Otherwise, the empty whitespace lines follow PEP-8 coding conventions in Python.</p>
B7D9EG6C9
2018-04-19 14:08:25
[cucumber/cucumber] jenisys commented on #342: tag-expressions for python
<p>That is a convenience function that allows you to compare the token enum-value with its string representation, like:</p>
<p><code>token.matches("or")
</code></p>
<p>NOTE: Seems to be currently not used in the code.</p>
B7D9EG6C9
2018-04-19 14:20:26
[cucumber/cucumber] jenisys commented on #342: tag-expressions for python
<p>OTHERWISE PRIMARY CONCERN: Why so many additional lines ?
That is primarily development and release infrastructure for Python. For example, the Makefile based approach cannot be used on Windows platform (or only if some undocumented assumptions are met). If you have time, we can look over it and I explain it to you.</p>
<p>NOTE: Short answer, that is normally what I do as development infrastructure when I maintain and release a python package.</p>
B7D9EG6C9
2018-05-10 13:02:28
[cucumber/cucumber] mpkorstanje commented on #385: datatable: implemented js datatable
<p>I added this in Java so I could hide the constructor. This makes it easier to add in other <code>create</code> methods later on that do some preprocessing before constructing the table. </p>
<p>You can't do this in JS so I think create is not needed.</p>
B7D9EG6C9
2018-05-10 13:05:27
B7D9EG6C9
2018-06-02 10:31:24
B7D9EG6C9
2018-06-12 11:47:33
[cucumber/cucumber] mpkorstanje commented on #408: datatable: Automatic list conversion
<p>To the best of my knowledge <code>objectMapper.convertValue("someThing", SingleArgConstructor.class);</code> works. So you use Jackson here as well. I believe it is also a bit more flexible as it will check for methods name "fromString" as well.</p>
<p>It simplifies the documentation (refer to Jackons)</p>
B7D9EG6C9
2018-06-12 15:27:23
B7D9EG6C9
2018-06-12 15:48:28
[cucumber/cucumber] mpkorstanje commented on #408: datatable: Automatic list conversion
<p>Can we add an implementation note here: Something along the lines of we're using Jackson, but don't rely on anything but getters/setter/constructor/fromString/valueOf methods?</p>
B7D9EG6C9
2018-06-12 16:06:23
B7D9EG6C9
2018-06-15 06:35:26
B7D9EG6C9
2018-06-15 06:36:24
B7D9EG6C9
2018-06-15 06:37:27
B7D9EG6C9
2018-06-15 06:43:29
B7D9EG6C9
2018-06-20 13:15:34
Aslak Hellesøy
2019-11-25 07:41:26
https://github.com/cucumber/cucumber/pull/804
<p><strong>Summary</strong></p>
<p>Support <code>DocString</code> and <code>DataTable</code> in <code>fake-cucumber</code></p>
<p><strong>Details</strong></p>
<p>For now, we just grab the <code>DocString#value</code> and the <code>DataTable</code>is transformed into a <code>string[][]</code>. This can be made pluggable later so users can provide their own converters (as Cucumber-JVM has)</p>
GitHub
2021-04-15 02:27:49
[cucumber/cucumber] Pull request opened by aurelien-reeves
<p><strong>Summary</strong></p>
<p>Fixes <a href="https://github.com/cucumber/cucumber/issues/1475">#1475</a></p>
<p>I moved the scripts away from the bin directory</p>
<p><strong>Types of changes</strong></p>
<p>• Bug fix (non-breaking change which fixes an issue).</p>
<p><strong>Checklist:</strong></p>
<p>☐ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-15 04:08:00
[cucumber/cucumber] Review on #
Looks good, thanks <a href="https://github.com/vincent-psarga">@vincent-psarga</a>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-15 07:41:41
[cucumber/cucumber] Issue opened by aurelien-reeves
<p>As we can see in the following CI builds, the <code>make</code> command does barely nothing by itself in <code>gherkin/python</code>.</p>
<p>• <a href="https://app.circleci.com/pipelines/github/cucumber/cucumber/5872/workflows/99353b13-28a5-47db-80f3-301d19150d00/jobs/219587">https://app.circleci.com/pipelines/github/cucumber/cucumber/5872/workflows/99353b13-28a5-47db-80f3-301d19150d00/jobs/219587</a>
• <a href="https://app.circleci.com/pipelines/github/cucumber/cucumber/5872/workflows/99353b13-28a5-47db-80f3-301d19150d00/jobs/219595">https://app.circleci.com/pipelines/github/cucumber/cucumber/5872/workflows/99353b13-28a5-47db-80f3-301d19150d00/jobs/219595</a></p>
<p>I encounter the same behavior on my development computer inside the docker container.</p>
<p><code>make default</code> however is working as expected.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-15 18:49:56
[cucumber/cucumber] Issue opened by mangjoze
<p>Built-in HTML report generates a blank HTML file</p>
<p><strong>Summary</strong></p>
<p>Running my test using cucumber's built in html report, it generates a blank html file. viewing the source shows a bunch of css tags and license.</p>
<p>On terminal, if I type <em>cucumber -t <a href="https://github.com/wip">@wip</a> --format html --out report.html.</em> This runs my test but it generates a blank report.html. I'm not sure if I need to install a 3rd party formatter or an additional ruby gem to cater to this.</p>
<p><strong>Expected Behavior</strong></p>
<p>I'm expecting it to have the html report where the feature file steps and screenshots are visible.</p>
<p><strong>Your Environment</strong></p>
<p>• Version used: cucumber 6.0.0
• Operating System and version: Windows 10<br />
ruby 2.7.0</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-16 00:56:10
[cucumber/cucumber] Comment on #1482
Hi <a href="https://github.com/mangjoze">@mangjoze</a>
You shouldn't need any 3rd party plugins. Can you try running it using `bundle` and can you also try running it with the cucumber version set to a 5.x version (We just released 6.0 recently)
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-16 00:57:59
[cucumber/cucumber] Comment on #1482
Also as you've created duplicate issues. I'll close this one off. And we can continue the conversation on the one on the cucumber-ruby repo.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-16 00:58:00
[cucumber/cucumber] Issue closed by luke-hill
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-16 13:47:24
[cucumber/cucumber] Issue opened by mohrmz
<p>I have created specflow.json and set language like :<br />
{<br />
"language": {<br />
"feature": "fa"<br />
}<br />
}</p>
<p>but when i use ویژگی as feature i got some errors :</p>
<p>Severity Code Description Project File Line Suppression State<br />
Error (2:1): expected: #EOF, #Language, #TagLine, #FeatureLine, #Comment, #Empty, got 'وِیژگی : تست'</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-18 02:24:48
[cucumber/cucumber] Issue closed by brasmusson
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-18 02:35:01
[cucumber/cucumber] Comment on #1481
The problem was introduced in <a href="https://github.com/cucumber/cucumber/pull/1458">#1458</a>.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-18 07:40:33
[cucumber/cucumber] Comment on #1480
The problem is not having python files in the bin directory per se (`bin/gherkin` was a python file in previous releases), the problem is to have both a directory `gherkin` with an file `__init__.py` and a file `gherkin.py` (outside the `gherkin` directory)
`import gherkin`
in python can either mean, import from the directory named `gherkin` with a file `__init__.py`, or import from the file `gherkin.py`. If the paths are setup so that the `import gherkin` finds the file `gherkin.py` first, things will go wrong since then the imports in `gherkin.py` for things located in the directory `gherkin` will not be found (like `gherkin.stream.gherkin_events`) since `gherkin` already have been set up to mean `gherkin.py`.
It is only the files that are part of the release that can cause these kinds of problem, therefore nothing have to be done with `bin/gherkin_generate_tokens.py` (`setup.py` defines which files that are part of the release).
When the released `gherkin-python` is installed the file `bin/gherkin` typically ends up in `~/.local/bin`, whereas the files in the `gherkin` directory typically ends up in `~/.local/lib/<python version>/site-packages/gherkin/`, so the files in `bin` (that are released) cannot make assumptions on where the `gherkin` directory are.
To both have `bin/gherkin` be able to different python version using different executable names (in out docker image, the python-2 executable is `usr/bin/python2` and the python-3 executable is `usr/bin/python3`) and get rid of `bin/gherkin.py`, I think the most sensible thing is to move `bin/gherkin.py` to `gherkin/__main__.py` and let `bin/gherkin` execute `gherkin-python` as a module (that is using the `-m` option to the python executable).
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-18 13:26:12
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
TypeStrong/ts-loader
<strong><a href="https://togithub.com/TypeStrong/ts-loader/blob/master/CHANGELOG.md#v900"><code>v9.0.0</code></a></strong></p>
<p><a href="https://togithub.com/TypeStrong/ts-loader/compare/v8.1.0...v9.0.0">Compare Source</a></p>
<p>Breaking changes:</p>
<p>• minimum webpack version: 5
• minimum node version: 12</p>
<p>Changes:</p>
<p>• <a href="https://togithub.com/TypeStrong/ts-loader/pull/1251">webpack 5 migration</a> - thanks <a href="https://togithub.com/johnnyreilly">@johnnyreilly</a>, <a href="https://togithub.com/jonwallsten">@jonwallsten</a>, <a href="https://togithub.com/sokra">@sokra</a>, <a href="https://togithub.com/appzuka">@appzuka</a>, <a href="https://togithub.com/alexander-akait">@alexander-akait</a></p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/cucumber">here</a>.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-19 00:48:31
[cucumber/cucumber] Comment on #1480
Thank you for your analysis <a href="https://github.com/brasmusson">@brasmusson</a>
That makes much more sense.
Anything else required to merge that PR?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-19 04:44:54
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
tj/commander.js
<strong><a href="https://togithub.com/tj/commander.js/blob/master/CHANGELOG.md#%E2%80%8B720-2021-03-26|<code">https://togithub.com/tj/commander.js/blob/master/CHANGELOG.md#%E2%80%8B720-2021-03-26|<code</a>v7.2.0</code>></strong></p>
<p><a href="https://togithub.com/tj/commander.js/compare/v7.1.0...v7.2.0">Compare Source</a></p>
<p><strong>Added</strong></p>
<p>• TypeScript typing for <code>parent</code> property on <code>Command</code> ([#1475])
• TypeScript typing for <code>.attributeName()</code> on <code>Option</code> ([#1483])
• support information in package ([#1477])</p>
<p><strong>Changed</strong></p>
<p>• improvements to error messages, README, and tests
• update dependencies</p>
<p><strong><a href="https://togithub.com/tj/commander.js/blob/master/CHANGELOG.md#%E2%80%8B710-2021-02-15|<code">https://togithub.com/tj/commander.js/blob/master/CHANGELOG.md#%E2%80%8B710-2021-02-15|<code</a>v7.1.0</code>></strong></p>
<p><a href="https://togithub.com/tj/commander.js/compare/v7.0.0...v7.1.0">Compare Source</a></p>
<p><strong>Added</strong></p>
<p>• support for named imports from ECMAScript modules ([#1440])
• add <code>.cjs</code> to list of expected script file extensions ([#1449])
• allow using option choices and variadic together ([#1454])</p>
<p><strong>Fixed</strong></p>
<p>• replace use of deprecated <code>process.mainModule</code> ([#1448])
• regression for legacy <code>command('**')</code> and call when command line includes options ([#1464])
• regression for <code>on('command:**', ...)</code> and call when command line includes unknown options ([#1464])
• display best error for combination of unknown command and unknown option (i.e. unknown command) ([#1464])</p>
<p><strong>Changed</strong></p>
<p>• make TypeScript typings tests stricter ([#1453])
• improvements to README and tests</p>
<p><strong><a href="https://togithub.com/tj/commander.js/blob/master/CHANGELOG.md#%E2%80%8B700-2021-01-15|<code">https://togithub.com/tj/commander.js/blob/master/CHANGELOG.md#%E2%80%8B700-2021-01-15|<code</a>v7.0.0</code>></strong></p>
<p><a href="https://togithub.com/tj/commander.js/compare/v6.2.1...v7.0.0">Compare Source</a></p>
<p><strong>Added</strong></p>
<p>• <code>.enablePositionalOptions()</code> to let program and subcommand reuse same option ([#1427])
• <code>.passThroughOptions()</code> to pass options through to other programs without needing <code>--</code> ([#1427])
• <code>.allowExcessArguments(false)</code> to show an error message if there are too many command-arguments on command line for the action handler ([#1409])
• <code>.configureOutput()</code> to modify use of stdout and stderr or customise display of errors ([#1387])
• use <code>.addHelpText()</code> to add text before or after the built-in help, for just current command or also for all subcommands ([#1296])
• enhance Option class ([#1331])
• allow hiding options from help
• allow restricting option arguments to a list of choices
• allow setting how default value is shown in help
• <code>.createOption()</code> to support subclassing of automatically created options (like <code>.createCommand()</code>) ([#1380])
• refactor the code generating the help into a separate public Help class ([#1365])
• support sorting subcommands and options in help
• support specifying wrap width (columns)
• allow subclassing Help class
• allow configuring Help class without subclassing</p>
<p><strong>Changed</strong></p>
<p>• <em>Breaking:</em> options are stored safely by default, not as properties on the command ([#1409])
• this especially affects accessing options on program, use <code>program.opts()</code>
• revert behaviour with <code>.storeOptionsAsProperties()</code>
• <em>Breaking:</em> action handlers are passed options and command separately ([#1409])
• deprecated callback parameter to <code>.help()</code> and <code>.outputHelp()</code> (removed from README) ([#1296])
• <em>Breaking:</em> errors now displayed using <code>process.stderr.write()</code> instead of <code>console.error()</code>
• deprecate <code>.on('--help')</code> (removed from README) ([#1296])
• initialise the command description to empty string (previously undefined) ([#1365])
• document and annotate deprecated routines ([#1349])</p>
<p><strong>Fixed</strong></p>
<p>• wrapping bugs in help ([#1365])
• first line of command description was wrapping two characters early
• pad width calculation was not including help option and help command
• pad width calculation was including hidden options and commands
• improve backwards compatibility for custom command event listeners ([#1403])</p>
<p><strong>Deleted</strong></p>
<p>• <em>Breaking:</em> <code>.passCommandToAction()</code> ([#1409])
• no longer needed as action handler is passed options and command
• <em>Breaking:</em> "extra arguments" parameter to action handler ([#1409])
• if being used to detect excess arguments, there is now an error available by setting <code>.allowExcessArguments(false)</code></p>
<p><strong>Migration Tips</strong></p>
<p>The biggest change is the parsed option values. Previously the options were stored by default as properties on the command object, and now the options are stored separately.</p>
<p>If you wish to restore the old behaviour and get running quickly you can call <code>.storeOptionsAsProperties()</code>.<br />
To allow you to move to the new code patterns incrementally, the action handler will be passed the command <em>twice</em>,<br />
to match the new "options" and "command" parameters (see below).</p>
<p><strong>program options</strong></p>
<p>Use the <code>.opts()</code> method to access the options. This is available on any command but is used most with the program.</p>
<pre><code>program.option('-d, --debug');
program.parse();
// Old code before Commander 7
if (program.debug) console.log(`Program name is ${program.name()}`);
</code></pre>
<pre><code>// New code
const options = program.opts();
if (options.debug) console.log(`Program name is ${program.name()}`);
</code></pre>
<p><strong>action handler</strong></p>
<p>The action handler gets passed a parameter for each command-argument you declared. Previously by default the next parameter was the command object with the options as properties. Now the next two parameters are instead the options and the command. If you<br />
only accessed the options there may be no code changes required.</p>
<pre><code>program
.command('compress &lt;filename&gt;')
.option('-t, --trace')
// Old code before Commander 7
.action((filename, cmd)) =&gt; {
if (cmd.trace) console.log(`Command name is ${cmd.name()}`);
});
</code></pre>
<pre><code> // New code
.action((filename, options, command)) =&gt; {
if (options.trace) console.log(`Command name is ${command.name()}`);
});
</code></pre>
<p>If you already set <code>.storeOptionsAsProperties(false)</code> you may still need to adjust your code.</p>
<pre><code>program
.command('compress &lt;filename&gt;')
.storeOptionsAsProperties(false)
.option('-t, --trace')
// Old code before Commander 7
.action((filename, command)) =&gt; {
if (command.opts().trace) console.log(`Command name is ${command.name()}`);
});
</code></pre>
<pre><code> // New code
.action((filename, options, command)) =&gt; {
if (command.opts().trace) console.log(`Command name is ${command.name()}`);
});
</code></pre>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/cucumber">here</a>.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-19 04:45:35
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
facebook/react
<strong><a href="https://togithub.com/facebook/react/blob/master/CHANGELOG.md#%E2%80%8B1702-March-22-2021|<code">https://togithub.com/facebook/react/blob/master/CHANGELOG.md#%E2%80%8B1702-March-22-2021|<code</a>v17.0.2</code>></strong></p>
<p><a href="https://togithub.com/facebook/react/compare/v17.0.1...v17.0.2">Compare Source</a></p>
<p><strong>React DOM</strong></p>
<p>• Remove an unused dependency to address the <a href="https://developer.chrome.com/blog/enabling-shared-array-buffer/"><code>SharedArrayBuffer</code> cross-origin isolation warning</a>. (<a href="https://togithub.com/koba04">@koba04</a> and <a href="https://togithub.com/bvaughn">@bvaughn</a> in <a href="https://togithub.com/facebook/react/pull/20831">#20831</a>, <a href="https://togithub.com/facebook/react/pull/20832">#20832</a>, and <a href="https://togithub.com/facebook/react/pull/20840">#20840</a>)</p>
<p><strong><a href="https://togithub.com/facebook/react/blob/master/CHANGELOG.md#%E2%80%8B1701-October-22-2020|<code">https://togithub.com/facebook/react/blob/master/CHANGELOG.md#%E2%80%8B1701-October-22-2020|<code</a>v17.0.1</code>></strong></p>
<p><a href="https://togithub.com/facebook/react/compare/v17.0.0...v17.0.1">Compare Source</a></p>
<p><strong>React DOM</strong></p>
<p>• Fix a crash in IE11. (<a href="https://togithub.com/gaearon">@gaearon</a> in <a href="https://togithub.com/facebook/react/pull/20071">#20071</a>)</p>
<p><strong><a href="https://togithub.com/facebook/react/blob/master/CHANGELOG.md#%E2%80%8B1700-October-20-2020|<code">https://togithub.com/facebook/react/blob/master/CHANGELOG.md#%E2%80%8B1700-October-20-2020|<code</a>v17.0.0</code>></strong></p>
<p><a href="https://togithub.com/facebook/react/compare/v16.13.1...v17.0.0">Compare Source</a></p>
<p>Today, we are releasing React 17!</p>
<p><strong><a href="https://reactjs.org/blog/2020/10/20/react-v17.html">Learn more about React 17 and how to update to it on the official React blog.</a></strong></p>
<p><strong>React</strong></p>
<p>• Add <code>react/jsx-runtime</code> and <code>react/jsx-dev-runtime</code> for the <a href="https://babeljs.io/blog/2020/03/16/7.9.0#a-new-jsx-transform-11154-https-githubcom-babel-babel-pull-11154">new JSX transform</a>. (<a href="https://togithub.com/lunaruan">@lunaruan</a> in <a href="https://togithub.com/facebook/react/pull/18299">#18299</a>)
• Build component stacks from native error frames. (<a href="https://togithub.com/sebmarkbage">@sebmarkbage</a> in <a href="https://togithub.com/facebook/react/pull/18561">#18561</a>)
• Allow to specify <code>displayName</code> on context for improved stacks. (<a href="https://togithub.com/eps1lon">@eps1lon</a> in <a href="https://togithub.com/facebook/react/pull/18224">#18224</a>)
• Prevent <code>'use strict'</code> from leaking in the UMD bundles. (<a href="https://togithub.com/koba04">@koba04</a> in <a href="https://togithub.com/facebook/react/pull/19614">#19614</a>)
• Stop using <code><a href="http://fb.me">fb.me</a></code> for redirects. (<a href="https://togithub.com/cylim">@cylim</a> in <a href="https://togithub.com/facebook/react/pull/19598">#19598</a>)</p>
<p><strong>React DOM</strong></p>
<p>• Delegate events to roots instead of <code>document</code>. (<a href="https://togithub.com/trueadm">@trueadm</a> in <a href="https://togithub.com/facebook/react/pull/18195">#18195</a> and <a href="https://togithub.com/facebook/react/pulls?q=is%3Apr+author%3Atrueadm+modern+event+is%3Amerged|others">https://togithub.com/facebook/react/pulls?q=is%3Apr+author%3Atrueadm+modern+event+is%3Amerged|others</a>)
• Clean up all effects before running any next effects. (<a href="https://togithub.com/bvaughn">@bvaughn</a> in <a href="https://togithub.com/facebook/react/pull/17947">#17947</a>)
• Run <code>useEffect</code> cleanup functions asynchronously. (<a href="https://togithub.com/bvaughn">@bvaughn</a> in <a href="https://togithub.com/facebook/react/pull/17925">#17925</a>)
• Use browser <code>focusin</code> and <code>focusout</code> for <code>onFocus</code> and <code>onBlur</code>. (<a href="https://togithub.com/trueadm">@trueadm</a> in <a href="https://togithub.com/facebook/react/pull/19186">#19186</a>)
• Make all <code>Capture</code> events use the browser capture phase. (<a href="https://togithub.com/trueadm">@trueadm</a> in <a href="https://togithub.com/facebook/react/pull/19221">#19221</a>)
• Don't emulate bubbling of the <code>onScroll</code> event. (<a href="https://togithub.com/gaearon">@gaearon</a> in <a href="https://togithub.com/facebook/react/pull/19464">#19464</a>)
• Throw if <code>forwardRef</code> or <code>memo</code> component returns <code>undefined</code>. (<a href="https://togithub.com/gaearon">@gaearon</a> in <a href="https://togithub.com/facebook/react/pull/19550">#19550</a>)
• Remove event pooling. (<a href="https://togithub.com/trueadm">@trueadm</a> in <a href="https://togithub.com/facebook/react/pull/18969">#18969</a>)
• Stop exposing internals that won’t be needed by React Native Web. (<a href="https://togithub.com/necolas">@necolas</a> in <a href="https://togithub.com/facebook/react/pull/18483">#18483</a>)
• Attach all known event listeners when the root mounts. (<a href="https://togithub.com/gaearon">@gaearon</a> in <a href="https://togithub.com/facebook/react/pull/19659">#19659</a>)
• Disable <code>console</code> in the second render pass of DEV mode double render. (<a href="https://togithub.com/sebmarkbage">@sebmarkbage</a> in <a href="https://togithub.com/facebook/react/pull/18547">#18547</a>)
• Deprecate the undocumented and misleading <code>ReactTestUtils.SimulateNative</code> API. (<a href="https://togithub.com/gaearon">@gaearon</a> in <a href="https://togithub.com/facebook/react/pull/13407">#13407</a>)
• Rename private field names used in the internals. (<a href="https://togithub.com/gaearon">@gaearon</a> in <a href="https://togithub.com/facebook/react/pull/18377">#18377</a>)
• Don't call User Timing API in development. (<a href="https://togithub.com/gaearon">@gaearon</a> in <a href="https://togithub.com/facebook/react/pull/18417">#18417</a>)
• Disable console during the repeated render in Strict Mode. (<a href="https://togithub.com/sebmarkbage">@sebmarkbage</a> in <a href="https://togithub.com/facebook/react/pull/18547">#18547</a>)
• In Strict Mode, double-render components without Hooks too. (<a href="https://togithub.com/eps1lon">@eps1lon</a> in <a href="https://togithub.com/facebook/react/pull/18430">#18430</a>)
• Allow calling <code>ReactDOM.flushSync</code> during lifecycle methods (but warn). (<a href="https://togithub.com/sebmarkbage">@sebmarkbage</a> in <a href="https://togithub.com/facebook/react/pull/18759">#18759</a>)
• Add the <code>code</code> property to the keyboard event objects. (<a href="https://togithub.com/bl00mber">@bl00mber</a> in <a href="https://togithub.com/facebook/react/pull/18287">#18287</a>)
• Add the <code>disableRemotePlayback</code> property for <code>video</code> elements. (<a href="https://togithub.com/tombrowndev">@tombrowndev</a> in <a href="https://togithub.com/facebook/react/pull/18619">#18619</a>)
• Add the <code>enterKeyHint</code> property for <code>input</code> elements. (<a href="https://togithub.com/eps1lon">@eps1lon</a> in <a href="https://togithub.com/facebook/react/pull/18634">#18634</a>)
• Warn when no <code>value</code> is provided to <code>&lt;Context.Provider&gt;</code>. (<a href="https://togithub.com/charlie1404">@charlie1404</a> in <a href="https://togithub.com/facebook/react/pull/19054">#19054</a>)
• Warn when <code>memo</code> or <code>forwardRef</code> components return <code>undefined</code>. (<a href="https://togithub.com/bvaughn">@bvaughn</a> in <a href="https://togithub.com/facebook/react/pull/19550">#19550</a>)
• Improve the error message for invalid updates. (<a href="https://togithub.com/JoviDeCroock">@JoviDeCroock</a> in <a href="https://togithub.com/facebook/react/pull/18316">#18316</a>)
• Exclude forwardRef and memo from stack frames. (<a href="https://togithub.com/sebmarkbage">@sebmarkbage</a> in <a href="https://togithub.com/facebook/react/pull/18559">#18559</a>)
• Improve the error message when switching between controlled and uncontrolled inputs. (<a href="https://togithub.com/vcarl">@vcarl</a> in <a href="https://togithub.com/facebook/react/pull/17070">#17070</a>)
• Keep <code>onTouchStart</code>, <code>onTouchMove</code>, and <code>onWheel</code> passive. (<a href="https://togithub.com/gaearon">@gaearon</a> in <a href="https://togithub.com/facebook/react/pull/19654">#19654</a>)
• Fix <code>setState</code> hanging in development inside a closed iframe. (<a href="https://togithub.com/gaearon">@gaearon</a> in <a href="https://togithub.com/facebook/react/pull/19220">#19220</a>)
• Fix rendering bailout for lazy components with <code>defaultProps</code>. (<a href="https://togithub.com/jddxf">@jddxf</a> in <a href="https://togithub.com/facebook/react/pull/18539">#18539</a>)
• Fix a false positive warning when <code>dangerouslySetInnerHTML</code> is <code>undefined</code>. (<a href="https://togithub.com/eps1lon">@eps1lon</a> in <a href="https://togithub.com/facebook/react/pull/18676">#18676</a>)
• Fix Test Utils with non-standard <code>require</code> implementation. (<a href="https://togithub.com/just-boris">@just-boris</a> in <a href="https://togithub.com/facebook/react/pull/18632">#18632</a>)
• Fix <code>onBeforeInput</code> reporting an incorrect <code>event.type</code>. (<a href="https://togithub.com/eps1lon">@eps1lon</a> in <a href="https://togithub.com/facebook/react/pull/19561">#19561</a>)
• Fix <code>event.relatedTarget</code> reported as <code>undefined</code> in Firefox. (<a href="https://togithub.com/claytercek">@claytercek</a> in <a href="https://togithub.com/facebook/react/pull/19607">#19607</a>)
• Fix "unspecified error" in IE11. (<a href="https://togithub.com/hemakshis">@hemakshis</a> in <a href="https://togithub.com/facebook/react/pull/19664">#19664</a>)
• Fix rendering into a shadow root. (<https://togithub.c…</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-19 04:46:20
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
uuidjs/uuid
<strong><a href="https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B832-httpsgithubcomuuidjsuuidcomparev831v832-2020-12-08|<code">https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B832-httpsgithubcomuuidjsuuidcomparev831v832-2020-12-08|<code</a>v8.3.2</code>></strong></p>
<p><a href="https://togithub.com/uuidjs/uuid/compare/v8.3.1...v8.3.2">Compare Source</a></p>
<p><strong><a href="https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B831-httpsgithubcomuuidjsuuidcomparev830v831-2020-10-04|<code">https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B831-httpsgithubcomuuidjsuuidcomparev830v831-2020-10-04|<code</a>v8.3.1</code>></strong></p>
<p><a href="https://togithub.com/uuidjs/uuid/compare/v8.3.0...v8.3.1">Compare Source</a></p>
<p><strong><a href="https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B830-httpsgithubcomuuidjsuuidcomparev820v830-2020-07-27|<code">https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B830-httpsgithubcomuuidjsuuidcomparev820v830-2020-07-27|<code</a>v8.3.0</code>></strong></p>
<p><a href="https://togithub.com/uuidjs/uuid/compare/v8.2.0...v8.3.0">Compare Source</a></p>
<p><strong>Features</strong></p>
<p>• add parse/stringify/validate/version/NIL APIs (<a href="https://togithub.com/uuidjs/uuid/issues/479">#479</a>) (<a href="https://togithub.com/uuidjs/uuid/commit/0e6c10ba1bf9517796ff23c052fc0468eedfd5f4">0e6c10b</a>), closes <a href="https://togithub.com/uuidjs/uuid/issues/475">#475</a> <a href="https://togithub.com/uuidjs/uuid/issues/478">#478</a> <a href="https://togithub.com/uuidjs/uuid/issues/480">#480</a> <a href="https://togithub.com/uuidjs/uuid/issues/481">#481</a> <a href="https://togithub.com/uuidjs/uuid/issues/180">#180</a></p>
<p><strong><a href="https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B820-httpsgithubcomuuidjsuuidcomparev810v820-2020-06-23|<code">https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B820-httpsgithubcomuuidjsuuidcomparev810v820-2020-06-23|<code</a>v8.2.0</code>></strong></p>
<p><a href="https://togithub.com/uuidjs/uuid/compare/v8.1.0...v8.2.0">Compare Source</a></p>
<p><strong>Features</strong></p>
<p>• improve performance of v1 string representation (<a href="https://togithub.com/uuidjs/uuid/issues/453">#453</a>) (<a href="https://togithub.com/uuidjs/uuid/commit/0ee0b67c37846529c66089880414d29f3ae132d5">0ee0b67</a>)
• remove deprecated v4 string parameter (<a href="https://togithub.com/uuidjs/uuid/issues/454">#454</a>) (<a href="https://togithub.com/uuidjs/uuid/commit/88ce3ca0ba046f60856de62c7ce03f7ba98ba46c">88ce3ca</a>), closes <a href="https://togithub.com/uuidjs/uuid/issues/437">#437</a>
• support jspm (<a href="https://togithub.com/uuidjs/uuid/issues/473">#473</a>) (<a href="https://togithub.com/uuidjs/uuid/commit/e9f2587a92575cac31bc1d4ae944e17c09756659">e9f2587</a>)</p>
<p><strong>Bug Fixes</strong></p>
<p>• prepare package exports for webpack 5 (<a href="https://togithub.com/uuidjs/uuid/issues/468">#468</a>) (<a href="https://togithub.com/uuidjs/uuid/commit/8d6e6a5f8965ca9575eb4d92e99a43435f4a58a8">8d6e6a5</a>)</p>
<p><strong><a href="https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B810-httpsgithubcomuuidjsuuidcomparev800v810-2020-05-20|<code">https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B810-httpsgithubcomuuidjsuuidcomparev800v810-2020-05-20|<code</a>v8.1.0</code>></strong></p>
<p><a href="https://togithub.com/uuidjs/uuid/compare/v8.0.0...v8.1.0">Compare Source</a></p>
<p><strong>Features</strong></p>
<p>• improve v4 performance by reusing random number array (<a href="https://togithub.com/uuidjs/uuid/issues/435">#435</a>) (<a href="https://togithub.com/uuidjs/uuid/commit/bf4af0d711b4d2ed03d1f74fd12ad0baa87dc79d">bf4af0d</a>)
• optimize V8 performance of bytesToUuid (<a href="https://togithub.com/uuidjs/uuid/issues/434">#434</a>) (<a href="https://togithub.com/uuidjs/uuid/commit/e156415448ec1af2351fa0b6660cfb22581971f2">e156415</a>)</p>
<p><strong>Bug Fixes</strong></p>
<p>• export package.json required by react-native and bundlers (<a href="https://togithub.com/uuidjs/uuid/issues/449">#449</a>) (<a href="https://togithub.com/uuidjs/uuid/commit/be1c8fe9a3206c358e0059b52fafd7213aa48a52">be1c8fe</a>), closes <a href="https://togithub.com/ai/nanoevents/issues/44#issuecomment-602010343">ai/nanoevents#44</a> <a href="https://togithub.com/uuidjs/uuid/issues/444">#444</a></p>
<p><strong><a href="https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B800-httpsgithubcomuuidjsuuidcomparev703v800-2020-04-29|<code">https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B800-httpsgithubcomuuidjsuuidcomparev703v800-2020-04-29|<code</a>v8.0.0</code>></strong></p>
<p><a href="https://togithub.com/uuidjs/uuid/compare/v7.0.3...v8.0.0">Compare Source</a></p>
<p><strong>⚠️ BREAKING CHANGES</strong></p>
<p>• For native ECMAScript Module (ESM) usage in Node.js only named exports are exposed, there is no more default export.
<pre><code>-import uuid from 'uuid';
-console.log(uuid.v4()); // -&gt; 'cd6c3b08-0adc-4f4b-a6ef-36087a1c9869'
+import { v4 as uuidv4 } from 'uuid';
+uuidv4(); // ⇨ '9b1deb4d-3b7d-4bad-9bdd-2b0d7b3dcb6d'
</code></pre>
• Deep requiring specific algorithms of this library like <code>require('uuid/v4')</code>, which has been deprecated in <code>uuid@7</code>, is no longer supported.</p>
<pre><code>Instead use the named exports that this module exports.
For ECMAScript Modules (ESM):
<pre><code>-import uuidv4 from 'uuid/v4';
+import { v4 as uuidv4 } from 'uuid';
uuidv4();
</code></pre>
For CommonJS:
<pre><code>-const uuidv4 = require('uuid/v4');
+const { v4: uuidv4 } = require('uuid');
uuidv4();
</code></pre>
</code></pre>
<p><strong>Features</strong></p>
<p>• native Node.js ES Modules (wrapper approach) (<a href="https://togithub.com/uuidjs/uuid/issues/423">#423</a>) (<a href="https://togithub.com/uuidjs/uuid/commit/2d9f590ad9701d692625c07ed62f0a0f91227991">2d9f590</a>), closes <a href="https://togithub.com/uuidjs/uuid/issues/245">#245</a> <a href="https://togithub.com/uuidjs/uuid/issues/419">#419</a> <a href="https://togithub.com/uuidjs/uuid/issues/342">#342</a>
• remove deep requires (<a href="https://togithub.com/uuidjs/uuid/issues/426">#426</a>) (<a href="https://togithub.com/uuidjs/uuid/commit/daf72b84ceb20272a81bb5fbddb05dd95922cbba">daf72b8</a>)</p>
<p><strong>Bug Fixes</strong></p>
<p>• add CommonJS syntax example to README quickstart section (<a href="https://togithub.com/uuidjs/uuid/issues/417">#417</a>) (<a href="https://togithub.com/uuidjs/uuid/commit/e0ec8402c7ad44b7ef0453036c612f5db513fda0">e0ec840</a>)</p>
<p><strong><a href="https://togithub.com/uuidjs/uuid/compare/v7.0.2...v7.0.3">7.0.3</a> (2020-03-31)</strong>
<strong>Bug Fixes</strong></p>
<p>• make deep require deprecation warning work in browsers (<a href="https://togithub.com/uuidjs/uuid/issues/409">#409</a>) (<a href="https://togithub.com/uuidjs/uuid/commit/4b71107d8c0d2ef56861ede6403fc9dc35a1e6bf">4b71107</a>), closes <a href="https://togithub.com/uuidjs/uuid/issues/408">#408</a></p>
<p><strong><a href="https://togithub.com/uuidjs/uuid/compare/v7.0.1...v7.0.2">7.0.2</a> (2020-03-04)</strong>
<strong>Bug Fixes</strong></p>
<p>• make access to msCrypto consistent (<a href="https://togithub.com/uuidjs/uuid/issues/393">#393</a>) (<a href="https://togithub.com/uuidjs/uuid/commit/8bf2a20f3565df743da7215eebdbada9d2df118c">8bf2a20</a>)
• simplify link in deprecation warning (<a href="https://togithub.com/uuidjs/uuid/issues/391">#391</a>) (<a href="https://togithub.com/uuidjs/uuid/commit/bb2c8e4e9f4c5f9c1eaaf3ea59710c633cd90cb7">bb2c8e4</a>)
• update links to match content in readme (<a href="https://togithub.com/uuidjs/uuid/issues/386">#386</a>) (<a href="https://togithub.com/uuidjs/uuid/commit/44f2f86e9d2bbf14ee5f0f00f72a3db1292666d4">44f2f86</a>)</p>
<p><strong><a href="https://togithub.com/uuidjs/uuid/compare/v7.0.0...v7.0.1">7.0.1</a> (2020-02-25)</strong>
<strong>Bug Fixes</strong></p>
<p>• clean up esm builds for node and browser (<a href="https://togithub.com/uuidjs/uuid/issues/383">#383</a>) (<a href="https://togithub.com/uuidjs/uuid/commit/59e6a49e7ce7b3e8fb0f3ee52b9daae72af467dc">59e6a49</a>)
• provide browser versions independent from module system (<a href="https://togithub.com/uuidjs/uuid/issues/380">#380</a>) (<a href="https://togithub.com/uuidjs/uuid/commit/4344a22e7aed33be8627eeaaf05360f256a21753">4344a22</a>), closes <a href="https://togithub.com/uuidjs/uuid/issues/378">#378</a></p>
<p><strong><a href="https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B703-httpsgithubcomuuidjsuuidcomparev702v703-2020-03-31|<code">https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B703-httpsgithubcomuuidjsuuidcomparev702v703-2020-03-31|<code</a>v7.0.3</code>></strong></p>
<p><a href="https://togithub.com/uuidjs/uuid/compare/v7.0.2...v7.0.3">Compare Source</a></p>
<p><strong><a href="https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B702-httpsgithubcomuuidjsuuidcomparev701v702-2020-03-04|<code">https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B702-httpsgithubcomuuidjsuuidcomparev701v702-2020-03-04|<code</a>v7.0.2</code>></strong></p>
<p><a href="https://togithub.com/uuidjs/uuid/compare/v7.0.1...v7.0.2">Compare Source</a></p>
<p><strong><a href="https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B701-httpsgithubcomuuidjsuuidcomparev700v701-2020-02-25|<code">https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B701-httpsgithubcomuuidjsuuidcomparev700v701-2020-02-25|<code</a>v7.0.1</code>></strong></p>
<p><a href="https://togithub.com/uuidjs/uuid/compare/v7.0.0...v7.0.1">Compare Source</a></p>
<p><strong><a href="https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B700-httpsgithubcomuuidjsuuidcomparev340v700-2020-02-24|<code">https://togithub.com/uuidjs/uuid/blob/master/CHANGELOG.md#%E2%80%8B700-httpsgithubcomuuidjsuuidcomparev340v700-2020-02-24|<code</a>v7.0.0</code>></strong></p>
<p><a href="https://togithub.com/uuidjs/uuid/compare/v3.4.0...v7.0.0">Compare Source</a></p>
<p><strong>⚠️ BREAKING CHANGES</strong></p>
<p>• The default export, which used to be the v4() method but which was already discouraged in v3.x of this library, has been removed.
• Explicitly note that deep imports of the different uuid version functions are deprecated and no longer encouraged and that ECMAScript module named imports should be used instead. Emit a deprecation warning for people who deep-require the different algorithm variants.
• Remove builtin support for insecure random number generators …</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-19 05:20:55
[cucumber/cucumber] Issue opened by davidjgoss
<p><strong>Summary</strong></p>
<p>From <code>react/javascript</code>, running the <code>storybook</code> script to do local development with storybook fails.</p>
<p><strong>Expected Behavior</strong></p>
<p>Storybook runs normally.</p>
<p><strong>Current Behavior</strong></p>
<pre><code>/Users/davidgoss/Projects/cucumber/node_modules/html-webpack-plugin/lib/webpack5/file-watcher-api.js:13
mainCompilation.fileSystemInfo.createSnapshot(
^
TypeError: Cannot read property 'createSnapshot' of undefined
at /Users/davidgoss/Projects/cucumber/node_modules/html-webpack-plugin/lib/webpack5/file-watcher-api.js:13:36
at new Promise (&lt;anonymous&gt;)
at Object.createSnapshot (/Users/davidgoss/Projects/cucumber/node_modules/html-webpack-plugin/lib/webpack5/file-watcher-api.js:12:10)
at /Users/davidgoss/Projects/cucumber/node_modules/html-webpack-plugin/lib/cached-child-compiler.js:219:35
</code></pre>
<p><strong>Possible Solution</strong></p>
<p>This kind of thing often happens when multiple webpack versions are in <code>node_modules</code> and a plugin or loader picks up the wrong one because of how the hoisting has come out. Looks like it may be the case here due to storybook addons. From <code>npm ls webpack</code>:</p>
<p>```
{
"version": "0.0.0",
"name": "@cucumber/cucumber-monorepo",
"problems": [
"invalid: webpack@5.30.0 /Users/davidgoss/Projects/cucumber/node<em>modules/webpack"
],
"dependencies": {
"@cucumber/html-formatter": {
"version": "13.0.0",
"resolved": "file:../../html-formatter/javascript",
"dependencies": {
"babel-loader": {
"version": "8.2.2",
"dependencies": {
"webpack": {
"version": "5.30.0"
}
}
},
"ts-loader": {
"version": "8.1.0",
"dependencies": {
"webpack": {
"version": "5.30.0"
}
}
},
"webpack-cli": {
"version": "4.6.0",
"dependencies": {
"@webpack-cli/configtest": {
"version": "1.0.2",
"dependencies": {
"webpack": {
"version": "5.30.0"
}
}
},
"webpack": {
"version": "5.30.0"
}
}
},
"webpack": {
"version": "5.30.0",
"dependencies": {
"terser-webpack-plugin": {
"version": "5.1.1",
"dependencies": {
"webpack": {
"version": "5.30.0"
}
}
}
}
}
}
},
"@cucumber/react": {
"version": "13.0.0",
"resolved": "file:../../react/javascript",
"dependencies": {
"@storybook/addon-essentials": {
"version": "6.2.3",
"resolved": "<a href="https://registry.npmjs.org/@storybook/addon-essentials/-/addon-essentials-6.2.3.tgz">https://registry.npmjs.org/@storybook/addon-essentials/-/addon-essentials-6.2.3.tgz</a>",
"dependencies": {
"@storybook/addon-docs": {
"version": "6.2.3",
"resolved": "<a href="https://registry.npmjs.org/@storybook/addon-docs/-/addon-docs-6.2.3.tgz">https://registry.npmjs.org/@storybook/addon-docs/-/addon-docs-6.2.3.tgz</a>",
"dependencies": {
"@storybook/builder-webpack4": {
"version": "6.2.3",
"resolved": "<a href="https://registry.npmjs.org/@storybook/builder-webpack4/-/builder-webpack4-6.2.3.tgz">https://registry.npmjs.org/@storybook/builder-webpack4/-/builder-webpack4-6.2.3.tgz</a>",
"dependencies": {
"css-loader": {
"version": "3.6.0",
"resolved": "<a href="https://registry.npmjs.org/css-loader/-/css-loader-3.6.0.tgz">https://registry.npmjs.org/css-loader/-/css-loader-3.6.0.tgz</a>",
"dependencies": {
"webpack": {
"version": "5.30.0"
}
}
},
"dotenv-webpack": {
"version": "1.8.0",
"resolved": "<a href="https://registry.npmjs.org/dotenv-webpack/-/dotenv-webpack-1.8.0.tgz">https://registry.npmjs.org/dotenv-webpack/-/dotenv-webpack-1.8.0.tgz</a>",
"dependencies": {
"webpack": {
"version": "5.30.0",
"invalid": true,
"problems": [
"invalid: webpack@5.30.0 /Users/davidgoss/Projects/cucumber/node</em>modules/webpack"
]
}
}
},
"html-webpack-plugin": {
"version": "4.5.2",
"resolved": "<a href="https://registry.npmjs.org/html-webpack-plugin/-/html-webpack-plugin-4.5.2.tgz">https://registry.npmjs.org/html-webpack-plugin/-/html-webpack-plugin-4.5.2.tgz</a>",
"dependencies": {
"webpack": {
"version": "5.30.0"
}
}
},
"postcss-loader": {
"version": "4.2.0",
"resolved": "<a href="https://registry.npmjs.org/postcss-loader/-/postcss-loader-4.2.0.tgz">https://registry.npmjs.org/postcss-loader/-/postcss-loader-4.2.0.tgz</a>",
"dependencies": {
"webpack": {
"version": "5.30.0"
}
}
},
"style-loader": {
"version": "1.3.0",
"resolved": "<a href="https://registry.npmjs.org/style-loader/-/style-loader-1.3.0.tgz">https://registry.npmjs.org/style-loader/-/style-loader-1.3.0.tgz</a>",
"dependencies": {
"webpack": {
"version": "4.46.0"
}
}
},
"terser-webpack-plugin": {
"version": "3.1.0",
"resolved": "<a href="https://registry.npmjs.org/terser-webpack-plugin/-/terser-webpack-plugin-3.1.0.tgz">https://registry.npmjs.org/terser-webpack-plugin/-/terser-webpack-plugin-3.1.0.tgz</a>",
"dependencies": {
"webpack": {
"version": "5.30.0"
}
}
},
"webpack-dev-middleware": {
"version": "3.7.3",
"resolved": "<a href="https://registry.npmjs.org/webpack-dev-middleware/-/webpack-dev-middleware-3.7.3.tgz">https://registry.npmjs.org/webpack-dev-middleware/-/webpack-dev-middleware-3.7.3.tgz</a>",
"dependencies": {
"webpack": {
"version": "5.30.0"
}
}
},
"webpack-filter-warnings-plugin": {
"version": "1.2.1",
"resolved": "<a href="https://registry.npmjs.org/webpack-filter-warnings-plugin/-/webpack-filter-warnings-plugin-1.2.1.tgz">https://registry.npmjs.org/webpack-filter-warnings-plugin/-/webpack-filter-warnings-plugin-1.2.1.tgz</a>",
"dependencies": {
"webpack": {
"version": "5.30.0",
"invalid": true,
"problems": [
"invalid: webpack@5.30.0 /Users/davidgoss/Projects/cucumber/node_modules/webpack"
]
}
}
},
"webpack": {
"version": "4.46.0",
"resolved": "<a href="https://registry.npmjs.org/webpack/-/webpack-4.46.0.tgz">https://registry.npmjs.org/webpack/-/webpack-4.46.0.tgz</a>",
"dependencies": {
"terser-webpack-plugin": {
"version": "1.4.5",
"resolved": "<a href="https://registry.npmjs.org/terser-webpack-plugin/-/terser-webpack-plugin-1.4.5.tgz">https://registry.npmjs.org/terser-webpack-plugin/-/terser-webpack-plugin-1.4.5.tgz</a>",
"dependencies": {
"webpack": {
"version": "4.46.0"
}
}
}
}
}
}
},
"webpack": {
"version": "5.30.0"
}
}
},
"webpack": {
"version": "5.30.0"
}
}
},
"@storybook/builder-webpack5": {
"version": "6.2.3",
…</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-19 06:31:20
[cucumber/cucumber] Pull request opened by aurelien-reeves
<p><strong>Summary</strong></p>
<p>Fixes <a href="https://github.com/cucumber/cucumber-js/issues/1621">cucumber/cucumber-js#1621</a></p>
<p>json-formatter fails with examples with tags.<br />
Other formatters are working well with tagged examples.</p>
<p><strong>Types of changes</strong></p>
<p>• Bug fix (non-breaking change which fixes an issue).</p>
<p><strong>Checklist:</strong></p>
<p>☐ I've added tests for my code.
☐ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-19 09:50:23
[cucumber/cucumber] Comment on #1483
The Persian keyword for `Feature` is `وِیژگی` (<a href="https://github.com/cucumber/cucumber/blob/81ff8d94c9a74d3bb3bbf492c17ef9f316a18ca6/gherkin/gherkin-languages.json#L1174">source</a>)
All keywords except step keywords have to be followed immediately by a `:` without a space. Do you still get the same parse error if you remove the space?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-19 10:11:28
[cucumber/cucumber] Comment on #1488
Specifically for `npm ls webpack@4`:
```
@cucumber/cucumber-monorepo@0.0.0 /Users/davidgoss/Projects/cucumber
└─┬ @cucumber/react@13.0.0 -> /Users/davidgoss/Projects/cucumber/react/javascript
├─┬ @storybook/addon-essentials@6.2.3
│ └─┬ @storybook/addon-docs@6.2.3
│ └─┬ @storybook/builder-webpack4@6.2.3
│ ├─┬ style-loader@1.3.0
│ │ └── webpack@4.46.0 deduped
│ └─┬ webpack@4.46.0
│ └─┬ terser-webpack-plugin@1.4.5
│ └── webpack@4.46.0 deduped
├─┬ @storybook/builder-webpack5@6.2.3
│ └─┬ @storybook/core-common@6.2.3
│ └─┬ webpack@4.46.0
│ └─┬ terser-webpack-plugin@1.4.5
│ └── webpack@4.46.0 deduped
└─┬ @storybook/react@6.2.3
├─┬ @storybook/core@6.2.3
│ └─┬ @storybook/core-server@6.2.3
│ ├─┬ style-loader@1.3.0
│ │ └── webpack@4.46.0 deduped
│ └─┬ webpack@4.46.0
│ └─┬ terser-webpack-plugin@1.4.5
│ └── webpack@4.46.0 deduped
└─┬ webpack@4.46.0
└─┬ terser-webpack-plugin@1.4.5
└── webpack@4.46.0 deduped
```
Confirmed Storybook is using webpack 4 - will try and fix that.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-19 11:54:29
[cucumber/cucumber] Comment on #1483
> The Persian keyword for `Feature` is `وِیژگی` (<a href="https://github.com/cucumber/cucumber/blob/81ff8d94c9a74d3bb3bbf492c17ef9f316a18ca6/gherkin/gherkin-languages.json#L1174">source</a>)
>
> All keywords except step keywords have to be followed immediately by a `:` without a space. Do you still get the same parse error if you remove the space?
Thank you. I tried. I do not have parser error now, but when I want to generate the step definitions, I have an Object reference not set to an instance of an object. and Code highlighting is not working
feature file content:
وِیژگی:تعریف کاربر
سناریو:تعریف نام کاربری جدید
با فرض :من میخواهم نام کاربری جدید اضافه کنم
هنگامی :یک نام کاربری جدید اضافه می کنم
آنگاه :آنگاه باید نام کاربری جدید اضافه شود
<a href="https://user-images.githubusercontent.com/34716164/115288209-6baa6f80-a166-11eb-8e9c-5129bb4751ce.png">Screenshot 2021-04-19 232012</a>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-19 12:39:19
[cucumber/cucumber] Issue closed by aslakhellesoy
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-19 12:39:19
[cucumber/cucumber] Comment on #1483
Glad it worked! I’m closing this issue.
The new issue you’re seeing is unrelated to Cucumber or the Gherkin parser. Please open an issue with SpecFlow.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-19 14:09:38
[cucumber/cucumber] Pull request closed by renovate[bot]
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-19 14:09:41
[cucumber/cucumber] Pull request closed by renovate[bot]
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-19 14:09:46
[cucumber/cucumber] Pull request closed by renovate[bot]
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-20 05:05:46
[cucumber/cucumber] Issue closed by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-20 05:05:46
[cucumber/cucumber] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-20 05:10:19
[cucumber/cucumber] Pull request ready for review by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-20 06:47:05
[cucumber/cucumber] Comment on #1479
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> before merging that PR, maybe taking a look at the storybooks for cucumber-react would be welcome as the update of react markdown has breaking changes
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-20 06:47:25
[cucumber/cucumber] Review on #
LGTM!
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-20 06:48:06
[cucumber/cucumber] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-20 06:48:46
[cucumber/cucumber] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-20 07:05:24
[cucumber/cucumber] Comment on #1414
ping <a href="https://github.com/mpkorstanje">@mpkorstanje</a> - what do you think we should use instead of jackson?
We definitely need a library that can serialise objects to JSON.
I don't think we need deserialisation, so we could use a simple library that only does serialisation. I've looked at <a href="https://github.com/fabienrenaud/java-json-benchmark">https://github.com/fabienrenaud/java-json-benchmark</a>, and not a single one of them has serialisation only (which increases the attack-vector surface).
Shall we just go with a shaded `com.eclipsesource.minimal-json` as we use in `create-meta`?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-20 09:32:31
[cucumber/cucumber] Comment on #1462
Idea:
• Final message stream does _not_ include messages only used by Cucumber's execution
• Pickles
• etc
We can add an extra property to the `TestStepFinished` message. Something like this...
```
type TestStepFinished = {
documentStepReferences: DocumentStepReference[]
}
type DocumentStepReference = {
astNodeId: string
uri: string
lineNumber: number
}
```
By denormalising this way, consuming messages becomes a lot easier, and we can probably scrap the `Query` module altogether.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-21 01:14:35
[cucumber/cucumber] Comment on #1488
<a href="https://github.com/davidjgoss">@davidjgoss</a> I took a look as part of <a href="https://github.com/cucumber/cucumber/pull/1479">#1479</a>
We have actually already updated our storybook configuration to use webpack 5.
I've been able to make it work by removing the dependency to `@storybook/addon-essentials`
I am trying also to update storybook as the latest is 6.2.8
I'll try to create a small project to reproduce it and to open an issue - or fill an existing one - with a minimal reproduction example.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-21 01:26:29
[cucumber/cucumber] Comment on #1479
That last commit should fix <a href="https://github.com/cucumber/cucumber/issues/1488">#1488</a>
Cc. <a href="https://github.com/davidjgoss">@davidjgoss</a>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-21 01:31:41
[cucumber/cucumber] Review on #
LGTM, thanks for fixing storybook!
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-21 01:51:23
[cucumber/cucumber] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-21 01:53:30
[cucumber/cucumber] Comment on #1476
Those old version of cucumber are expected, to generated specific form of json
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-21 01:53:31
[cucumber/cucumber] Pull request closed by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-21 01:54:43
[cucumber/cucumber] Comment on #1476
*Renovate Ignore Notification*
As this PR has been closed unmerged, Renovate will ignore this upgrade and you will not receive PRs for _any_ future 6.x releases. However, if you upgrade to 6.x manually then Renovate will then reenable updates for minor and patch updates automatically.
If this PR was closed by mistake or you changed your mind, you can simply rename this PR and you will soon get a fresh replacement PR opened.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-21 02:43:31
[cucumber/cucumber] Comment on #1414
You want to turn abitrary POJOs into json right? `minimal-json` only supports turning json objects into json.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-21 03:05:27
[cucumber/cucumber] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-21 03:32:39
[cucumber/cucumber] Issue opened by aslakhellesoy
<p>Environment variables are documented here: <a href="https://devcenter.bitrise.io/builds/available-environment-variables/">https://devcenter.bitrise.io/builds/available-environment-variables/</a></p>
<p>To get started, add an entry to <a href="https://github.com/cucumber/cucumber/blob/master/create-meta/ciDict.json">https://github.com/cucumber/cucumber/blob/master/create-meta/ciDict.json</a> and run the tests...</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-21 03:36:20
[cucumber/cucumber] Comment on #1414
> You want to turn abitrary POJOs into json right? minimal-json only supports turning json objects into json.
Ah yes, forgot about that. Yes, we need to turn POJOs into JSON.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-21 05:54:19
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
puppeteer/puppeteer
<strong><a href="https://togithub.com/puppeteer/puppeteer/blob/master/CHANGELOG.md#%E2%80%8B900-httpsgithubcompuppeteerpuppeteercomparev800v900-2021-04-21|<code">https://togithub.com/puppeteer/puppeteer/blob/master/CHANGELOG.md#%E2%80%8B900-httpsgithubcompuppeteerpuppeteercomparev800v900-2021-04-21|<code</a>v9.0.0</code>></strong></p>
<p><a href="https://togithub.com/puppeteer/puppeteer/compare/v8.0.0...v9.0.0">Compare Source</a></p>
<p><strong>⚠️ BREAKING CHANGES</strong></p>
<p>• <strong>filechooser:</strong> FileChooser.cancel() is now synchronous.</p>
<p><strong>Features</strong></p>
<p>• <strong>chromium:</strong> roll to Chromium 91.0.4469.0 (r869685) (<a href="https://togithub.com/puppeteer/puppeteer/issues/7110">#7110</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/715e7a8d62901d1c7ec602425c2fce8d8148b742">715e7a8</a>)
• <strong>launcher:</strong> fix installation error on Apple M1 chips (<a href="https://togithub.com/puppeteer/puppeteer/issues/7099">#7099</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/c239d9edc72d85697b4875c98fff3ec592848082">c239d9e</a>), closes <a href="https://togithub.com/puppeteer/puppeteer/issues/6622">#6622</a>
• <strong>network:</strong> request interception and caching compatibility (<a href="https://togithub.com/puppeteer/puppeteer/issues/6996">#6996</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/8695759a223bc1bd31baecb00dc28721216e4c6f">8695759</a>)
• <strong>page:</strong> emit the event after removing the Worker (<a href="https://togithub.com/puppeteer/puppeteer/issues/7080">#7080</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/e34a6d53183c3e1f63a375ba6a26bee0dcfcf542">e34a6d5</a>)
• <strong>types:</strong> improve type of predicate function (<a href="https://togithub.com/puppeteer/puppeteer/issues/6997">#6997</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/943477cc1eb4b129870142873b3554737d5ef252">943477c</a>), closes <a href="https://togithub.com/puppeteer//github.com/DefinitelyTyped/DefinitelyTyped/blob/c43191a8f7a7d2a47bbff0bc3a7d95ecc64d2269/types/puppeteer/index.d.ts/issues/L1883-L1885">/github.com/DefinitelyTyped/DefinitelyTyped/blob/c43191a8f7a7d2a47bbff0bc3a7d95ecc64d2269/types/puppeteer/index.d.ts#L1883-L1885</a>
• accept captureBeyondViewport as optional screenshot param (<a href="https://togithub.com/puppeteer/puppeteer/issues/7063">#7063</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/0e092d2ea0ec18ad7f07ad3507deb80f96086e7a">0e092d2</a>)
• <strong>page:</strong> add omitBackground option for page.pdf method (<a href="https://togithub.com/puppeteer/puppeteer/issues/6981">#6981</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/dc8ab6d8ca1661f8e56d329e6d9c49c891e8b975">dc8ab6d</a>)</p>
<p><strong>Bug Fixes</strong></p>
<p>• <strong>aria:</strong> fix parsing of ARIA selectors (<a href="https://togithub.com/puppeteer/puppeteer/issues/7037">#7037</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/4426135692ae3ee7ed2841569dd9375e7ca8286c">4426135</a>)
• <strong>page:</strong> fix mouse.click method (<a href="https://togithub.com/puppeteer/puppeteer/issues/7097">#7097</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/ba7c367de33ace7753fd9d8b8cc894b2c14ab6c2">ba7c367</a>), closes <a href="https://togithub.com/puppeteer/puppeteer/issues/6462">#6462</a> <a href="https://togithub.com/puppeteer/puppeteer/issues/3347">#3347</a>
• make <code>$</code> and <code>$$</code> selectors generic (<a href="https://togithub.com/puppeteer/puppeteer/issues/6883">#6883</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/b349c91e7df76630b7411d6645e649945c4609bd">b349c91</a>)
• type page event listeners correctly (<a href="https://togithub.com/puppeteer/puppeteer/issues/6891">#6891</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/866d34ee1122e89eab00743246676845bb065968">866d34e</a>)
• <strong>typescript:</strong> allow defaultViewport to be 'null' (<a href="https://togithub.com/puppeteer/puppeteer/issues/6942">#6942</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/e31e68dfa12dd50482b700472bc98876b9031829">e31e68d</a>), closes <a href="https://togithub.com/puppeteer/puppeteer/issues/6885">#6885</a>
• make screenshots work in puppeteer-web (<a href="https://togithub.com/puppeteer/puppeteer/issues/6936">#6936</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/5f24f608194fd4252da7b288461427cabc9dabb3">5f24f60</a>)
• <strong>filechooser:</strong> cancel is sync (<a href="https://togithub.com/puppeteer/puppeteer/issues/6937">#6937</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/2ba61e04e923edaac09c92315212552f2d4ce676">2ba61e0</a>)
• <strong>network:</strong> don't disable cache for auth challenge (<a href="https://togithub.com/puppeteer/puppeteer/issues/6962">#6962</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/1c2479a6cd4bd09a577175ffd31c40ca6f4279b8">1c2479a</a>)</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/cucumber">here</a>.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-21 14:57:48
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/cucumber">here</a>.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-21 23:07:55
[cucumber/cucumber] Issue opened by Roman-Sht
<p><strong>Summary</strong></p>
<p>I am trying to use the message formatter that creates ndjson files and convert it to JSON using Cucumber Standalone JSON Formatter. However, there is a problem that after the conversion to JSON, all the scenarios steps appear in the "after" section instead in the "steps section.</p>
<p><strong>Expected Behavior</strong></p>
<p>Steps should go under "steps" section in the JSON</p>
<p><strong>Current Behavior</strong></p>
<p>Steps are put under "after" section in the JSON</p>
<p><strong>Steps to Reproduce (for bugs)</strong></p>
<p>cat tests.ndjson | cucumber-json-formatter --format ndjson > stand<em>alone</em>output.json</p>
<p><strong>Your Environment</strong></p>
<p>ruby 2.7.2<br />
cucumber 5.1.3</p>
<p><a href="https://github.com/cucumber/cucumber/files/6355897/input.txt">input.txt</a></p>
<p><a href="https://github.com/cucumber/cucumber/files/6355901/stand_alone_output.txt">stand<em>alone</em>output.txt</a></p>
<p>Comment: had to change the file extensions to upload them. Convert the input file extension to .ndjson and the output file to .json to reproduce</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-22 01:13:10
[cucumber/cucumber] Comment on #1493
Would you have the possibility to share more regarding your project like step definitions and hooks? Or a minimal reproduction example?
From what I see in your input, it looks like at least one of your After hook (`features/support/hooks.rb:96`) is executed after each step. That may be the cause of your issue.
Without your step definitions and hooks, or better, a minimal reproduction example, we can not go further. The issue here is not with the json formatter. Your ndjson input seems wrong.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-22 01:14:38
[cucumber/cucumber] Issue closed by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-22 01:15:12
[cucumber/cucumber] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-22 01:40:30
[cucumber/cucumber] New release Release - cucumber-gherkin/v18.1.1 published by cukebot
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-22 01:48:38
[cucumber/cucumber] Comment on #1475
<a href="https://github.com/lhupfeldt">@lhupfeldt</a> gherkin 18.1.1 has been released and should fix that issue
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-22 01:54:28
[cucumber/cucumber] Comment on #1493
<a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> the supplied `input.txt` should be sufficient to reproduce this. Why do we need more information?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-22 01:59:04
[cucumber/cucumber] New release Release - cucumber-json-formatter/v17.0.2 published by cukebot
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-22 02:03:36
[cucumber/cucumber] Comment on #1493
> Your ndjson input seems wrong.
Well, it's what's produced by Cucumber-Ruby. Do you think there is something wrong about how Cucumber-Ruby emits messages <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> ?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-22 02:14:57
[cucumber/cucumber] New release Release - cucumber-json-formatter/v17.0.2 published by cukebot
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-22 02:43:42
[cucumber/cucumber] Comment on #1493
> <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> the supplied `input.txt` should be sufficient to reproduce this. Why do we need more information?
> > Your ndjson input seems wrong.
>
> Well, it's what's produced by Cucumber-Ruby. Do you think there is something wrong about how Cucumber-Ruby emits messages <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> ?
What I see in the `input.txt` is something that looks like the following:
• testCaseStarted
• testStepStarted of a hook
• testStepStarted of another hook
• testStepStarted for a step definition
• testStepStarted of a hook
• testStepStarted for a step definition
• testStepStarted of a hook
• testStepStarted for a step definition
• testStepStarted of a hook
• ... and it continues like that until the testCaseFinished
From what I understand here is that there is a hook which is executed before or after each step. I assumed that behavior was not possible
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-22 02:53:29
[cucumber/cucumber] Comment on #1493
Oh, I did not knew about Step Hooks
My mistake 😞
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-22 03:02:10
[cucumber/cucumber] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-22 07:50:46
[cucumber/cucumber] Comment on #1414
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> after the discussions today with <a href="https://github.com/gasparnagy">@gasparnagy</a> at the committers' Zoom, it struck me that it might be a good idea to write an <a href="https://github.com/npryce/adr-tools">ADR</a> as part of this PR. I'd be happy to write a draft if you like, or we could pair/ensemble on it?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-22 09:06:50
[cucumber/cucumber] Issue opened by jwrightecs
<p>gherkin-perl: Gherkin uses feature from perl5.12 but only requires perl5.10.1</p>
<p><strong>Summary</strong></p>
<p>When running gherkin on perl 5.10.1 (the minimum supported version), the "..." lines in Gherkin::Pickles::Compiler cause a syntax error, perl minimum version should be increased or '...' feature not used</p>
<p><strong>Expected Behavior</strong></p>
<p>'gherkin' command should parse and run and return ndjson formatted messages</p>
<p><strong>Current Behavior</strong></p>
<p>gherkin command fails with syntax error</p>
<pre><code>$ gherkin
syntax error at /home/gherkin/perl5/lib/perl5/Gherkin/Pickles/Compiler.pm line 35, near "..."
syntax error at /home/gherkin/perl5/lib/perl5/Gherkin/Pickles/Compiler.pm line 171, near "..."
Compilation failed in require at /home/gherkin/perl5/lib/perl5/Gherkin.pm line 9.
</code></pre>
<p><strong>Possible Solution</strong></p>
<p>One of</p>
<p>• Gherkin module should have 5.12 as the minimum version, and not be installable on 5.10.1, or
• Gherkin::Pickles::Compiler should use 'die "Unimplemented";' instead of '...', making it 5.10.1 compatible</p>
<p><strong>Steps to Reproduce (for bugs)</strong></p>
<ol><li>perlbrew use 5.10.1</li><li>cpanm Gherkin</li><li>gherkin</li>
</ol>
<p><strong>Context & Motivation</strong></p>
<p>Trying to use Gherkin to parse files on an old system (Centos 6, Perl 5.10.1)</p>
<p><strong>Your Environment</strong></p>
<p>• Version used: Gherkin-18.1.1
• Operating System and version: CentOS 6
• Link to your project:</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-22 12:18:24
[cucumber/cucumber] Comment on #1494
Thanks for the report! We'll get this fixed (back to 5.10.1) and organize a release.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-22 12:56:45
[cucumber/cucumber] Pull request opened by ehuelsmann
<p><strong>Summary</strong></p>
<p>Close <a href="https://github.com/cucumber/cucumber/issues/1494">#1494</a>: Reinstate Perl 5.10.1 compatibility</p>
<p><strong>Details</strong></p>
<p>The 'yada-yada' operator (triple-dot) was added to Perl 5.12. It's functionally equivalent to <code>die "Unimplemented";</code>. The rest of the code base is 5.10.1 compliant. These two (minor) issues do not seem like a good reason to increase the requirement.</p>
<p><strong>How Has This Been Tested?</strong></p>
<p><code>make test</code> in the docker development environment.</p>
<p><strong>Types of changes</strong></p>
<p>☑︎ Bug fix (non-breaking change which fixes an issue).</p>
<p><strong>Checklist:</strong></p>
<p>☐ I've added tests for my code.
☐ My change requires a change to the documentation.
☐ I have updated the documentation accordingly.
☐ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-22 15:30:06
[cucumber/cucumber] Issue closed by ehuelsmann
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-22 15:30:06
[cucumber/cucumber] Pull request merged by ehuelsmann
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-23 22:15:37
[cucumber/cucumber] Issue opened by tarkers
<p><strong>Summary</strong></p>
<p>javascript<br />
it returns <strong>TypeError: Cannot read property 'reset' of undefined</strong><br />
should i give the custom parser builder and tokermatcher but how to provide id?</p>
<pre><code>var Gherkin = require("gherkin");
var parser = new Gherkin.Parser();
var gherkinDocument = parser.parse("Feature: ..."); // get error at this line
var pickles = new Gherkin.Compiler().compile(gherkinDocument);
</code></pre>
<p><strong>Expected Behavior</strong></p>
<p>return gherkinDocument successfully</p>
<p><strong>Current Behavior</strong></p>
<pre><code>TypeError: Cannot read property 'reset' of undefined
</code></pre>
<p><strong>Your Environment</strong></p>
<p>• Version used:"@cucumber/gherkin": "18.1.1"
• Operating System and version: windows 10</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 00:33:30
[cucumber/cucumber] Comment on #1496
The gherkin parser needs an AstBuilder and a TokenMatcher. Cf. <a href="https://github.com/cucumber/cucumber/blob/master/gherkin/javascript/src/Parser.ts#L109">https://github.com/cucumber/cucumber/blob/master/gherkin/javascript/src/Parser.ts#L109</a>
You can take a look at AstBuilder here: <a href="https://github.com/cucumber/cucumber/blob/master/gherkin/javascript/src/AstBuilder.ts">https://github.com/cucumber/cucumber/blob/master/gherkin/javascript/src/AstBuilder.ts</a>
And TokenMatcher here: <a href="https://github.com/cucumber/cucumber/blob/master/gherkin/javascript/src/TokenMatcher.ts">https://github.com/cucumber/cucumber/blob/master/gherkin/javascript/src/TokenMatcher.ts</a>
May I ask: what are you trying to build? What are your needs?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 01:00:08
[cucumber/cucumber] Comment on #1414
> <a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> after the discussions today with <a href="https://github.com/gasparnagy">@gasparnagy</a> at the committers' Zoom, it struck me that it might be a good idea to write an <a href="https://github.com/npryce/adr-tools">ADR</a> as part of this PR. I'd be happy to write a draft if you like, or we could pair/ensemble on it?
Yes, an ADR would be great <a href="https://github.com/mattwynne">@mattwynne</a>!
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 01:02:45
[cucumber/cucumber] Comment on #1496
I want to parse the feature file into json file.
In python , i use *gherkin-official*
```
from gherkin.token_scanner import TokenScanner
from gherkin.parser import Parser
parser = Parser()
# text read from my feature file
ff = parser.parse(TokenScanner(text))
```
the ff data is the thing i want ,than i hope if there are also some function in javascript
i 've check the doc but i can''t figure out what i should do to get such data.
Seems it doesn't need AstBuilder or a TokenMatcher ?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 01:17:20
[cucumber/cucumber] Comment on #1496
Official implementation may be close, they are not identical.
As I said: try instanciating an AstBuilder and a TokenMatcher to properly instanciate the parser.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 01:48:18
[cucumber/cucumber] Comment on #1496
Ok , i'll give it a try ,thanks your help
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 02:23:29
[cucumber/cucumber] Comment on #1496
If you encounter any other difficulties, of course you can reopen that issue.
Also, feel free to ask the community: <a href="https://cucumber.io/docs/community/">https://cucumber.io/docs/community/</a>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 02:23:29
[cucumber/cucumber] Issue closed by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 02:26:00
[cucumber/cucumber] Comment on #1496
<a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> can you update the `README.md` please? It's still incorrect.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 02:26:00
[cucumber/cucumber] Issue reopened by aslakhellesoy
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 03:06:20
[cucumber/cucumber] Comment on #1496
I missed the fact that the code reported in the issue was actually a snippet from one of our readme.
I'll submit a PR soon. I am afraid we may have to update a lot of the snippets.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 03:15:04
[cucumber/cucumber] Comment on #1496
<a href="https://github.com/tarkers">@tarkers</a> in the PR <a href="https://github.com/cucumber/cucumber/pull/1497">#1497</a> you should find a better snippet.
Sorry for my previous messages. I missed that you have found your own snippet in a readme in the repo.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 03:17:45
[cucumber/cucumber] Review on #
LGTM!
It would be nice to provide default values so there isn't so much to set up for users of the library, but that can go in a separate issue.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 03:18:35
[cucumber/cucumber] Comment on #1497
Some context may still be missing, like some explanation regarding the IdGenerator for example.
Any idea how we could improve documentation about the usage of the gherkin library more than just a basic snippet in the readme? The question may actually concern all sub-projects.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 03:19:49
[cucumber/cucumber] Comment on #1497
> LGTM!
>
> It would be nice to provide default values so there isn't so much to set up for users of the library, but that can go in a separate issue.
I will dig deeper indeed.
I will also check snippets for other languages.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 03:27:13
[cucumber/cucumber] Comment on #249
At the moment, it looks like the gherkin parser is parsing empty scenarios as requested in the initial message of that issue.
cucumber-ruby is reporting the scenario as undefined.
cucumber-jvm is reporting the scenario as passed.
cucumber-js is reporting something like "2 scenarios (1 passed)" when no hooks are set. It is reporting it passed when hooks are executed.
So, what needs to be done now? Should we try to aligning cucumber-jvm and cucumber-js on cucumber-ruby? Should that be part of that same issue? Maybe we could close it and open new one in cucumber-jvm and cucumber-js?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 03:43:38
[cucumber/cucumber] Comment on #1496
<a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> thx, follow <a href="https://github.com/cucumber/cucumber/pull/1497">#1497</a> javascript code
```
var Gherkin = require("@cucumber/gherkin")
var Messages = require("@cucumber/messages")
var uuidFn = Messages.IdGenerator.uuid;
var builder = new Gherkin.AstBuilder(uuidFn);
var matcher = new Gherkin.TokenMatcher();
var parser = new Gherkin.Parser(builder, matcher);
var gherkinDocument = parser.parse("Feature: ...");
var pickles = Gherkin.compile(gherkinDocument, 'uri_of_the_feature.feature', uuidFn)
```
it works!
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 07:10:27
[cucumber/cucumber] Comment on #249
We'll open a PR with an empty scenario in the compatibility kit
The expected result should be the scenario to be undefined, and no hooks to be executed - like cucumber-ruby
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 07:10:27
[cucumber/cucumber] Issue closed by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 08:49:51
[cucumber/cucumber] Comment on #249
👍 let's make sure to link from this issue to the PR once it's open so people can follow the trail.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-26 08:50:22
[cucumber/cucumber] Comment on #249
Personally I would keep this issue open until the PR is merged but I don't feel strongly about that.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-27 01:27:43
[cucumber/cucumber] Comment on #249
<a href="https://github.com/cucumber/cucumber/pull/1498">#1498</a>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-27 09:35:07
[cucumber/cucumber] Pull request opened by hkosova
<p><strong>Summary</strong></p>
<p>This PR adds another Russian term for "Scenario outline" - "Шаблон сценария" (scenario template).</p>
<p><strong>Details</strong>
<strong>Motivation and Context</strong></p>
<p>Improve Russian translation.</p>
<p>I think "Шаблон сценария" conveys the meaning better than the existing translation.</p>
<p><strong>How Has This Been Tested?</strong></p>
<p>N/a</p>
<p><strong>Screenshots (if appropriate):</strong>
<strong>Types of changes</strong></p>
<p>☐ Bug fix (non-breaking change which fixes an issue).
☑︎ New feature (non-breaking change which adds functionality).
☐ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<p><strong>Checklist:</strong></p>
<p>☑︎ The change has been ported to Java.
☑︎ The change has been ported to Ruby.
☑︎ The change has been ported to JavaScript.
☑︎ The change has been ported to Go.
☑︎ The change has been ported to .NET.
☐ I've added tests for my code.
☐ My change requires a change to the documentation.
☐ I have updated the documentation accordingly.
☐ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-27 12:28:36
[cucumber/cucumber] Issue opened by aslansservant42
<p><strong>Summary</strong></p>
<p>Sonatype registers a High Vulnerability issue with io.cucumber : cucumber-core : 6.10.3</p>
<p><strong>Expected Behavior</strong></p>
<p>Threat Level 6, Problem Code <a href="https://github.com/advisories/GHSA-9mrv-456v-pf22">CVE-2020-28487</a><br />
Related to using vis-timeline before 7.4.4 library</p>
<p>See below for Sonatype's description of the error</p>
<p><strong>Possible Solution</strong></p>
<p>Upgrade vis-timeline to the latest and most secure version.</p>
<p><strong>Sonatype Documenation</strong>
<strong>DESCRIPTION FROM CVE</strong></p>
<p>This affects the package vis-timeline before 7.4.4. An attacker with the ability to control the items of a Timeline element can inject additional script code into the generated application.</p>
<p><strong>EXPLANATION</strong></p>
<p>The vis-timeline package is vulnerable to Cross-Site Scripting (XSS) attacks. The functions in the files listed below fail to sanitize certain user-supplied parameters before rendering their values in HTML. A remote attacker who can influence Timeline elements can exploit this vulnerability by crafting a request containing malicious JavaScript in any affected parameter. The attacker can then entice a victim into submitting the request or interacting with the injection if it is stored. This will result in script execution when the resulting response is rendered by the victim's browser.</p>
<p><strong>Vulnerable File(s) and Function(s):</strong></p>
<p>package/lib/shared/Configurator.js</p>
<p>_makeHeader()<br />
_makeLabel()<br />
_setupPopup()<br />
package/lib/shared/Popup.js</p>
<p>setText()<br />
package/lib/timeline/Timeline.js</p>
<p>constructor()<br />
package/lib/timeline/component/CustomTime.js</p>
<p>setCustomMarker()<br />
package/lib/timeline/component/DataAxis.js</p>
<p>_redrawLabel()<br />
_redrawTitle()<br />
package/lib/timeline/component/Group.js</p>
<p>setData()<br />
package/lib/timeline/component/Legend.js</p>
<p>function()<br />
package/lib/timeline/component/TimeAxis.js</p>
<p>_repaintMinorText()<br />
_repaintMajorText()<br />
package/lib/timeline/component/item/Item.js</p>
<p>_repaintOnItemUpdateTimeTooltip()<br />
_updateContents()<br />
package/dist/vis-timeline-graph2d.esm.js</p>
<p>package/esnext/esm/vis-timeline-graph2d.js</p>
<p>package/esnext/umd/vis-timeline-graph2d.js</p>
<p>package/peer/esm/vis-timeline-graph2d.js</p>
<p>package/peer/umd/vis-timeline-graph2d.js</p>
<p>package/standalone/esm/vis-timeline-graph2d.js</p>
<p>package/standalone/umd/vis-timeline-graph2d.js</p>
<p>package/dist/vis-timeline-graph2d.min.js</p>
<p>package/esnext/esm/vis-timeline-graph2d.min.js</p>
<p>package/esnext/umd/vis-timeline-graph2d.min.js</p>
<p>package/peer/esm/vis-timeline-graph2d.min.js</p>
<p>package/peer/umd/vis-timeline-graph2d.min.js</p>
<p>package/standalone/esm/vis-timeline-graph2d.min.js</p>
<p>package/standalone/umd/vis-timeline-graph2d.min.js</p>
<p><strong>DETECTION</strong></p>
<p>The application is vulnerable by using this component.</p>
<p><strong>RECOMMENDATION</strong></p>
<p>We recommend upgrading to a version of this component that is not vulnerable to this specific issue.</p>
<p>Note: If this component is included as a bundled/transitive dependency of another component, there may not be an upgrade path. In this instance, we recommend contacting the maintainers who included the vulnerable package. Alternatively, we recommend investigating alternative components or a potential mitigating control.</p>
<p><strong>ROOT CAUSE</strong></p>
<p>cucumber-core-6.10.3.jario/cucumber/core/plugin/timeline/vis.min.js[0.3.0, 5.0.0)<br />
ADVISORIES<br />
Project:<a href="https://github.com/<https://github.com/visjs/vis-timeline/issues/838%7Cvisjs/vis-timeline/issues/838">https://github.com/<https://github.com/visjs/vis-timeline/issues/838%7Cvisjs/vis-timeline/issues/838</a>><br />
Project:<a href="https://github.com/<https://github.com/visjs/vis-timeline/pull/840%7Cvisjs/vis-timeline/pull/840">https://github.com/<https://github.com/visjs/vis-timeline/pull/840%7Cvisjs/vis-timeline/pull/840</a>><br />
CVSS DETAILS<br />
CVE CVSS 3:6.8<br />
CVSS Vector:CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:U/C:H/I:L/A:L</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-27 13:44:02
[cucumber/cucumber] Comment on #904
This is still an issue in 2021. Is there any timeline of this bug being fixed?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-28 09:13:37
[cucumber/cucumber] Comment on #1499
Thanks! We also need an entry in the `CHANGELOG.md` file.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-28 16:25:20
[cucumber/cucumber] Pull request ready for review by aslakhellesoy
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-29 00:38:11
[cucumber/cucumber] Comment on #1414
The removal of "sudo" before installing npm globally was the thing to fix the CI? 😨
This is a big PR to review. Here's a big list of question 😅
• What should we focus on while reviewing?
• What are the next steps?
• When could it be merged?
• Should we wait for the PR in cucumber-js, cucumber-ruby and cucumber-jvm to be ready too?
• Should we wait for .Net and Elixir to be implemented?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-29 00:39:32
[cucumber/cucumber] Comment on #1497
Just a follow-up: I want to check the java snippet too before requesting for review/merging this.
I am currently in the process of setting-up a Java development environment 😅
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-29 00:46:20
[cucumber/cucumber] Comment on #1414
> The removal of "sudo" before installing npm globally was the thing to fix the CI? 😨
>
> This is a big PR to review. Here's a big list of question 😅
>
> • What should we focus on while reviewing?
Primarily finding breaking changes in APIs and data structures and update changelogs
> • What are the next steps?
Merge to master and continue working on Cucumber PRs
> • When could it be merged?
I think we can merge now, unless there are objections.
> • Should we wait for the PR in cucumber-js, cucumber-ruby and cucumber-jvm to be ready too?
I think we can work on those after this is merged.
> • Should we wait for .Net and Elixir to be implemented?
And Perl. I think we should disable them all (they already are). Waiting for them will cause too much delay. They can catch up at their own pace. We should create an issue for all 3 of them.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-29 00:55:03
[cucumber/cucumber] Comment on #1500
I did not find any reference to vis-timeline in the repo
We don't have a package, or gem, or anything which could match "cucumber-core : 6.10.3"
Would you have more info please?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-29 07:35:42
[cucumber/cucumber] davidjgoss approved aslakhellesoy's pull request
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-29 07:36:27
[cucumber/cucumber] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-29 07:38:09
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/cucumber">here</a>.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-29 07:38:20
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
joelittlejohn/jsonschema2pojo
<strong><a href="https://togithub.com/joelittlejohn/jsonschema2pojo/blob/master/CHANGELOG.md#%E2%80%8B111|<code">https://togithub.com/joelittlejohn/jsonschema2pojo/blob/master/CHANGELOG.md#%E2%80%8B111|<code</a>v1.1.1</code>></strong></p>
<p>• nested sibling references causes error <code>Path not present: definitions</code> only on <a href="https://www.jsonschema2pojo.org/">https://www.jsonschema2pojo.org/</a>, local is fine (<a href="https://togithub.com/joelittlejohn/jsonschema2pojo/issues/1237">#1237</a>)
• Provide a JDK9+ compatible annotations library for Android projects (<a href="https://togithub.com/joelittlejohn/jsonschema2pojo/issues/1218">#1218</a>)
• $comment field for property can be turned into JavaDoc (<a href="https://togithub.com/joelittlejohn/jsonschema2pojo/issues/1197">#1197</a>)
• UnnecessaryModifier PMD warning in generated code (private constructor for enums) (<a href="https://togithub.com/joelittlejohn/jsonschema2pojo/issues/1102">#1102</a>)
• When providing example json with a value great than Integer/MAX_VALUE, it should generate a long field in Java (<a href="https://togithub.com/joelittlejohn/jsonschema2pojo/issues/971">#971</a>)
• IllegalArgumentException while resolving reference to reference with # (<a href="https://togithub.com/joelittlejohn/jsonschema2pojo/issues/793">#793</a>)</p>
<p><strong><a href="https://togithub.com/joelittlejohn/jsonschema2pojo/blob/master/CHANGELOG.md#%E2%80%8B110|<code">https://togithub.com/joelittlejohn/jsonschema2pojo/blob/master/CHANGELOG.md#%E2%80%8B110|<code</a>v1.1.0</code>></strong></p>
<p>• Add <a href="https://togithub.com/Generated">@Generated</a> annotation to classes and includeGeneratedAnnotation config option (<a href="https://togithub.com/joelittlejohn/jsonschema2pojo/pull/1202">#1202</a>)
• <strong>Remove Jackson 1.x support (<a href="https://togithub.com/joelittlejohn/jsonschema2pojo/issues/1200">#1200</a>)</strong>
• <strong>Remove jsonschema2pojo-scalagen (<a href="https://togithub.com/joelittlejohn/jsonschema2pojo/issues/1199">#1199</a>)</strong>
• Enum generation ignores existingJavaType (<a href="https://togithub.com/joelittlejohn/jsonschema2pojo/issues/1182">#1182</a>)
• <strong>Require Gradle 5.6+ when using the Gradle plugin (and update examples) (<a href="https://togithub.com/joelittlejohn/jsonschema2pojo/pull/1131">#1131</a>)</strong>
• formatTypeMapping fails with primitive types (<a href="https://togithub.com/joelittlejohn/jsonschema2pojo/issues/1126">#1126</a>)
• Add private modifier to constructor-related configuration options so that they can be configured in the POM (<a href="https://togithub.com/joelittlejohn/jsonschema2pojo/pull/1118">#1118</a>)
• Using enum + javaType (without package name) generates enum in default package as a class, then fails to compile (<a href="https://togithub.com/joelittlejohn/jsonschema2pojo/issues/1103">#1103</a>)
• Fix raw type warnings when extending a builder (<a href="https://togithub.com/joelittlejohn/jsonschema2pojo/pull/1101">#1101</a>)
• Recursive schemas with relative paths can generate the same class twice with "__1" in the name (<a href="https://togithub.com/joelittlejohn/jsonschema2pojo/pull/1077">#1077</a>)
• Gradle shows deprecation warnings regarding Gradle 7.0 (<a href="https://togithub.com/joelittlejohn/jsonschema2pojo/issues/1070">#1070</a>)
• Add javax.annotation.processing.Generated for Java 9 (<a href="https://togithub.com/joelittlejohn/jsonschema2pojo/issues/835">#835</a>)</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/cucumber">here</a>.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-29 09:40:21
[cucumber/cucumber] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-29 09:41:07
[cucumber/cucumber] Pull request ready for review by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-29 10:48:39
[cucumber/cucumber] Comment on #1498
The current issue with the CI is:
• the `json-formatter` uses `cucumber-ruby 3` to generate a reference json
• `cucumber-ruby 3` did not managed empty scenarios properly. `cucumber-ruby 4` would
• `cucumber-ruby 4` resulting json has some data in the `match.location` attribute that cannot be used for reference
In order to go further with that PR, I suggest:
• we update `cucumber-ruby` in `json-formatter/ruby-testdata` to at least version 4
• we update the sanitizer scripts to make the resulting json looks like we want to
Any other idea?
Note: the extra data in `match.location` are those multiple line locations which comes from example tables:
```
"match": {
"location": "../../compatibility-kit/javascript/features/examples-tables/examples-tables.feature:27:9"
},
```
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-30 00:12:54
[cucumber/cucumber] Pull request merged by mpkorstanje
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-30 00:14:37
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/cucumber">here</a>.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-30 00:17:44
[cucumber/cucumber] Issue closed by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-30 00:17:45
[cucumber/cucumber] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-30 01:02:08
[cucumber/cucumber] Issue opened by laeubi
<https://github.com/cucumber/cucumber/issues/1504|#1504 Supply expected / actual in seperate fields>
<p>Currently one get the following message for a (failed) test<em>step</em>finished</p>
<pre><code>test_step_finished {
test_step_result {
status: FAILED
message: "org.opentest4j.AssertionFailedError: expected: &lt;8.0&gt; but was: &lt;3.0&gt;\n\tat org.junit.jupiter.api.AssertionUtils.fail(AssertionUtils.java:55)\n\tat org.junit.jupiter.api.AssertionUtils.failNotEqual(AssertionUtils.java:62)\n\tat org.junit.jupiter.api.AssertEquals.assertEquals(AssertEquals.java:182)\n\tat org.junit.jupiter.api.AssertEquals.assertEquals(AssertEquals.java:177)\n\tat org.junit.jupiter.api.Assertions.assertEquals(Assertions.java:1124)\n\tat io.cucumber.examples.java.RpnCalculatorSteps.the_result_is(io/cucumber/examples/java/basic_arithmetic.feature:27)\n"
duration {
nanos: 2984000
}
}
timestamp {
seconds: 1619766696
nanos: 398493000
}
test_step_id: "c2692c18-91e6-4a38-8cd8-c40ef71ca67e"
test_case_started_id: "22500d5c-fb17-464a-8f7e-8afa01342e58"
}
</code></pre>
<p>to prevent the consumer from parsing the message is would be good to have the 'expected: <8.0> but was: <3.0>' part split in different fields</p>
<p><strong>Describe the solution you'd like</strong></p>
<pre><code>test_step_finished {
test_step_result {
status: FAILED
expected: "&lt;8.0&gt;",
actual: "&lt;3.0&gt;",
message: "org.opentest4j.AssertionFailedError: expected: &lt;8.0&gt; but was: &lt;3.0&gt;\n\tat org.junit.jupiter.api.AssertionUtils.fail(AssertionUtils.java:55)\n\tat org.junit.jupiter.api.AssertionUtils.failNotEqual(AssertionUtils.java:62)\n\tat org.junit.jupiter.api.AssertEquals.assertEquals(AssertEquals.java:182)\n\tat org.junit.jupiter.api.AssertEquals.assertEquals(AssertEquals.java:177)\n\tat org.junit.jupiter.api.Assertions.assertEquals(Assertions.java:1124)\n\tat io.cucumber.examples.java.RpnCalculatorSteps.the_result_is(io/cucumber/examples/java/basic_arithmetic.feature:27)\n"
duration {
nanos: 2984000
}
}
timestamp {
seconds: 1619766696
nanos: 398493000
}
test_step_id: "c2692c18-91e6-4a38-8cd8-c40ef71ca67e"
test_case_started_id: "22500d5c-fb17-464a-8f7e-8afa01342e58"
}
</code></pre>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-30 01:09:14
[cucumber/cucumber] Comment on #1504
Thanks for clarification.
> The proposal would add these well fields to the message protocol when available.
Correctly, that would be good, to the tools consuming the messages don't need to know if it was java or some other runner...
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-30 01:20:11
[cucumber/cucumber] Comment on #1504
There is an implicit assumption here that an assertion is comparison between two values.
What is the type of `value` here? Looking at the opentest4j it looks like it can be objects of _any_ type: <a href="https://github.com/ota4j-team/opentest4j/blob/master/src/main/java/org/opentest4j/ValueWrapper.java">https://github.com/ota4j-team/opentest4j/blob/master/src/main/java/org/opentest4j/ValueWrapper.java</a>
So if we do this, the `actual` and `expected` fields could be `5`, `"5"`, `{"age": 5}`, `"2021-04-30T08:18:37.302Z"` etc.
I suppose the consumers (reporting tools) could render primitives (strings, numbers, dates) fine. Arrays of primitives could be rendered in a bullet list. But for deeply nested objects we'd have to display them as a tree.
Something to think about...
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-30 02:03:43
[cucumber/cucumber] Issue opened by enikao
<p><strong>Summary</strong></p>
<p>Comments on the same line as a tag cannot be retrieved from the list of comments.<br />
Applies to tags on both Features and Scenarios.</p>
<p>Example:</p>
<pre><code># I can find this comment
@myTag <b>#but</b> not this one
Feature: Make comments available through the parser
# comment to be found
@myOtherTag <b>#inaccessible</b> comment
Scenario: If a tag is followed by a comment, the comment should be accessible through the parser.
</code></pre>
<p><strong>Expected Behavior</strong></p>
<p>All comments, including the ones following a tag on the same line, should be present in GherkinDocument.getCommentsList().</p>
<p><strong>Current Behavior</strong></p>
<p>Comments after a tag on the same line are inaccessible via the parser.</p>
<p><strong>Context & Motivation</strong></p>
<p>I want to import the complete file contents.<br />
We often use comments after tags to simulate "tags with spaces", so these comments contain valuable information.</p>
<p><strong>Your Environment</strong></p>
<p>• Version used: io.cucumber.gherkin 17.0.2
• Operating System and version: Windows 10, JDK 11.0.7 amd64
• Link to your project: private</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-30 02:37:45
[cucumber/cucumber] Comment on #1498
The json report from cucumber-ruby seems actually wrong. Instead of resulting in an empty scenario, it results in a background
```
[
{
"description": " Scenario can be empty.\n Background and hooks are not executed.\n The state of the scenario is `undefined`",
"elements": [
{
"description": "",
"keyword": "Background",
"line": 6,
"name": "",
"type": "background"
}
],
"id": "minimal",
"keyword": "Feature",
"line": 1,
"name": "minimal",
"uri": "features/empty-scenario/empty-scenario.feature"
}
]
```
The result of the json-formatter when converting messages from fake-cucumber looks pretty good to me:
```
[
{
"description": " Scenario can be empty.\n Background and hooks are not executed.\n The state of the scenario is `undefined`",
"elements": [
{
"id": "minimal;empty",
"keyword": "Scenario",
"line": 9,
"name": "empty",
"steps": [],
"type": "scenario"
}
],
"id": "minimal",
"keyword": "Feature",
"line": 1,
"name": "minimal",
"uri": "features/empty-scenario/empty-scenario.feature"
}
]
```
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-30 06:49:57
[cucumber/cucumber] Comment on #1391
Discussed at open source day:
• Include tailwind in the cucumber-react and html-formatter stack so we can port some of the improved styling of the "chrome" around the report (meta, stats, search etc) over from Cucumber Reports
cc <a href="https://github.com/aslakhellesoy">@aslakhellesoy</a>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-30 06:55:23
[cucumber/cucumber] Comment on #1209
Jon Acker suggested looking at <a href="https://concordion.org/extensions/java/markdown/">https://concordion.org/extensions/java/markdown/</a>
Also suggested for tags:
```
`@foo` `@bar`
# Feature: hello
```
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-30 07:33:08
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/cucumber">here</a>.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-30 09:37:35
[cucumber/cucumber] Issue opened by davidjgoss
<p>Small step for inclusivity. Already been done on <a href="https://github.com/cucumber/cucumber-jvm">cucumber-jvm</a></p>
<p>☐ cucumber-js
☐ cucumber-ruby
☐ docs
☐ godog</p>
<p>cc <a href="https://github.com/mattwynne">@mattwynne</a></p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-30 10:08:22
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
syntax-tree/hast-util-sanitize
<strong><a href="https://togithub.com/syntax-tree/hast-util-sanitize/releases/4.0.0"><code>v4.0.0</code></a></strong></p>
<p><a href="https://togithub.com/syntax-tree/hast-util-sanitize/compare/3.0.2...4.0.0">Compare Source</a></p>
<p>• <a href="https://togithub.com/syntax-tree/hast-util-sanitize/commit/12f7f9b"><code>12f7f9b</code></a> Add JSDoc based types
• <a href="https://togithub.com/syntax-tree/hast-util-sanitize/commit/1076942"><code>1076942</code></a> Use ESM</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/cucumber">here</a>.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-30 22:55:25
[cucumber/cucumber] Comment on #1507
<a href="https://github.com/jaysonesmith">@jaysonesmith</a> can you help with godog?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-30 23:11:01
[cucumber/cucumber] Comment on #1507
For people with a local checkout, here's the steps you need to take:
```
# Switch to the "master" branch:
$ git checkout master
# Rename it to "main":
$ git branch -m master main
# Get the latest commits (and branches!) from the remote:
$ git fetch
# Remove the existing tracking connection with "origin/master":
$ git branch --unset-upstream
# Create a new tracking connection with the new "origin/main" branch:
$ git branch -u origin/main
```
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-30 23:12:43
[cucumber/cucumber] Comment on #1507
<a href="https://www.hanselman.com/blog/easily-rename-your-git-default-branch-from-master-to-main">https://www.hanselman.com/blog/easily-rename-your-git-default-branch-from-master-to-main</a>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-04-30 23:23:09
[cucumber/cucumber] Comment on #1507
> <a href="https://github.com/jaysonesmith">@jaysonesmith</a> can you help with godog?
Absolutely!
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-01 06:48:58
[cucumber/cucumber] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-01 09:49:40
[cucumber/cucumber] Pull request opened by aslakhellesoy
<p>Some leftovers that weren't removed in <a href="https://github.com/cucumber/cucumber/pull/1414">#1414</a></p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-01 14:42:01
[cucumber/cucumber] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-02 10:41:36
[cucumber/cucumber] Pull request opened by davidjgoss
<p><strong>Summary</strong></p>
<p>If <code>relativeTo</code> is provided in the options object for <code>fromPaths</code>, we'll calculate the <code>uri</code> for the <code>source</code> message (which is reused in subsequent ones) relative to that.</p>
<p><strong>Motivation and Context</strong></p>
<p>This is useful where e.g. we have a set of absolute paths for features which were found based on user input, but we don't want to show the full absolute path in reports etc - instead we want to show them relative to the root of the project we're running from (or something else in other edge cases where people use Gherkin directly).</p>
<p>See <a href="https://github.com/cucumber/cucumber-js/issues/1534#issuecomment-754716365">cucumber/cucumber-js#1534 (comment)</a></p>
<p><strong>Types of changes</strong></p>
<p>☐ Bug fix (non-breaking change which fixes an issue).
☑︎ New feature (non-breaking change which adds functionality).
☐ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<p><strong>Checklist:</strong></p>
<p>☐ The change has been ported to Java.
☐ The change has been ported to Ruby.
☐ The change has been ported to JavaScript.
☐ The change has been ported to Go.
☐ The change has been ported to .NET.
☑︎ I've added tests for my code.
☑︎ My change requires a change to the documentation.
☑︎ I have updated the documentation accordingly.
☑︎ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-02 12:26:32
[cucumber/cucumber] Comment on #1510
Can we link to ../gherkin-streams/javascript instead?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-02 12:26:32
[cucumber/cucumber] aslakhellesoy approved davidjgoss's pull request
Looks great!!
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-02 12:37:53
[cucumber/cucumber] Comment on #1510
Sure - I’ll make that an absolute URL so it’s useful for people hitting the readme via npm <a href="http://js.org">js.org</a>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-03 01:25:17
[cucumber/cucumber] Pull request merged by davidjgoss
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-03 23:14:28
[cucumber/cucumber] Issue opened by mpkorstanje
<p>The following is a valid gherkin document with two empty tables according to the gherkin/java 18.1.1 parser. However at a glance the tables do not <em>look</em> empty.</p>
<pre><code>Feature: Empty tables?
Scenario: This is valid Gherkin
Given a non-empty list
| hello
</code></pre>
<pre><code>Feature: Empty tables?
Scenario Outline: This is also valid Gherkin
Given a step &lt;hello&gt;
Examples:
| hello
</code></pre>
<p>For more complex tables this will confused users about what their tables actually contain. This can be especially insidious when tables are automatically transformed into POJO's and used as DTOs. It also has knock on effects on other parts of Cucumber. For example <a href="https://stackoverflow.com/questions/67361463/cucumber-doesnt-generate-steps-properly">in this SO question</a>, the user is confused about incorrect step definition generation caused by the ignoring the unfinished table cell.</p>
<p>It is essential that a users are not mislead about the contents of a table. So when encountering an unfinished table cell, i.e. any non-whitespace characters after the last <code>|</code> on a table row line I would expect the parser to throw an exception and explain the problem.</p>
<pre><code>import io.cucumber.gherkin.Gherkin;
import io.cucumber.messages.Messages;
import org.junit.jupiter.api.Test;
import java.util.UUID;
import static io.cucumber.gherkin.Gherkin.makeSourceEnvelope;
import static java.util.Collections.singletonList;
import static org.junit.jupiter.api.Assertions.assertEquals;
class FeatureParserTest {
@Test
void non_empty_table() {
String source = "" +
"Feature: Empty table?\n" +
" Scenario: This is valid Gherkin\n" +
" Given a non-empty list\n" +
" | hello \n";
Messages.PickleStepArgument.PickleTable pickleTable = Gherkin.fromSources(
singletonList(makeSourceEnvelope(source, "table.feature")),
false,
false,
true,
() -&gt; UUID.randomUUID().toString()
)
.map(Messages.Envelope::getPickle)
.map(pickle -&gt; pickle.getSteps(0))
.map(Messages.Pickle.PickleStep::getArgument)
.map(Messages.PickleStepArgument::getDataTable)
.findFirst()
.get();
assertEquals(1, pickleTable.getRowsList().size());
assertEquals(1, pickleTable.getRows(0).getCellsCount()); // fails, table is empty
}
@Test
void one_example() {
String source = "" +
"Feature: Single example?\n" +
" Scenario Outline: This is valid Gherkin\n" +
" Given a step &lt;hello&gt;\n" +
" Examples: \n" +
" | hello \n";
Messages.GherkinDocument.Feature.TableRow row = Gherkin.fromSources(
singletonList(makeSourceEnvelope(source, "table.feature")),
false,
true,
false,
() -&gt; UUID.randomUUID().toString()
)
.filter(Messages.Envelope::hasGherkinDocument)
.map(Messages.Envelope::getGherkinDocument)
.map(Messages.GherkinDocument::getFeature)
.map(feature -&gt; feature.getChildren(0))
.map(Messages.GherkinDocument.Feature.FeatureChild::getScenario)
.map(Messages.GherkinDocument.Feature.Scenario::getExamplesList)
.map(examples -&gt; examples.get(0))
.map(Messages.GherkinDocument.Feature.Scenario.Examples::getTableHeader)
.findFirst()
.get();
assertEquals(1, row.getCellsCount()); // fails, table is empty
}
}
</code></pre>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-04 11:23:24
[cucumber/cucumber] Pull request opened by OrhanTozan
<p><strong>Summary</strong>
<strong>Details</strong>
<strong>Motivation and Context</strong>
<strong>How Has This Been Tested?</strong>
<strong>Screenshots (if appropriate):</strong>
<strong>Types of changes</strong></p>
<p>☐ Bug fix (non-breaking change which fixes an issue).
☐ New feature (non-breaking change which adds functionality).
☐ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<p><strong>Checklist:</strong></p>
<p>☐ The change has been ported to Java.
☐ The change has been ported to Ruby.
☐ The change has been ported to JavaScript.
☐ The change has been ported to Go.
☐ The change has been ported to .NET.
☐ I've added tests for my code.
☐ My change requires a change to the documentation.
☐ I have updated the documentation accordingly.
☐ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-04 12:26:39
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
onsi/ginkgo
<strong><a href="https://togithub.com/onsi/ginkgo/releases/v1.16.2"><code>v1.16.2</code></a></strong></p>
<p><a href="https://togithub.com/onsi/ginkgo/compare/v1.16.1...v1.16.2">Compare Source</a></p>
<p><strong>1.16.2</strong>
<strong>Fixes</strong></p>
<p>• Deprecations can be suppressed by setting an <code>ACK_GINKGO_DEPRECATIONS=&lt;semver&gt;</code> environment variable.</p>
<p><strong><a href="https://togithub.com/onsi/ginkgo/releases/v1.16.1"><code>v1.16.1</code></a></strong></p>
<p><a href="https://togithub.com/onsi/ginkgo/compare/v1.16.0...v1.16.1">Compare Source</a></p>
<p><strong>Fixes</strong></p>
<p>• Supress --stream deprecation warning on windows (<a href="https://togithub.com/onsi/ginkgo/issues/793">#793</a>)</p>
<p><strong><a href="https://togithub.com/onsi/ginkgo/releases/v1.16.0"><code>v1.16.0</code></a></strong></p>
<p><a href="https://togithub.com/onsi/ginkgo/compare/v1.15.2...v1.16.0">Compare Source</a></p>
<p><strong>Features</strong></p>
<p>• Advertise Ginkgo 2.0. Introduce deprecations. [9ef1913]</p>
<pre><code>• Update README.md to advertise that Ginkgo 2.0 is coming.
• Backport the 2.0 DeprecationTracker and start alerting users
about upcoming deprecations.
</code></pre>
<p>• Add slim-sprig template functions to bootstrap/generate (<a href="https://togithub.com/onsi/ginkgo/issues/775">#775</a>) [9162b86]</p>
<p><strong>Fixes</strong></p>
<p>• Fix accidental reference to 1488 (<a href="https://togithub.com/onsi/ginkgo/issues/784">#784</a>) [9fb7fe4]</p>
<p><strong><a href="https://togithub.com/onsi/ginkgo/releases/v1.15.2"><code>v1.15.2</code></a></strong></p>
<p><a href="https://togithub.com/onsi/ginkgo/compare/v1.15.1...v1.15.2">Compare Source</a></p>
<p><strong>Fixes</strong></p>
<p>• ignore blank <code>-focus</code> and <code>-skip</code> flags (<a href="https://togithub.com/onsi/ginkgo/issues/780">#780</a>) [e90a4a0]</p>
<p><strong><a href="https://togithub.com/onsi/ginkgo/releases/v1.15.1"><code>v1.15.1</code></a></strong></p>
<p><a href="https://togithub.com/onsi/ginkgo/compare/v1.15.0...v1.15.1">Compare Source</a></p>
<p><strong>Fixes</strong></p>
<p>• reporters/junit: Use <code>system-out</code> element instead of <code>passed</code> (<a href="https://togithub.com/onsi/ginkgo/issues/769">#769</a>) [9eda305]</p>
<p><strong><a href="https://togithub.com/onsi/ginkgo/releases/v1.15.0"><code>v1.15.0</code></a></strong></p>
<p><a href="https://togithub.com/onsi/ginkgo/compare/v1.14.2...v1.15.0">Compare Source</a></p>
<p><strong>Features</strong></p>
<p>• Adds 'outline' command to print the outline of specs/containers in a file (<a href="https://togithub.com/onsi/ginkgo/issues/754">#754</a>) [071c369][6803cc3] [935b538][06744e8] [0c40583]
• Add support for using template to generate tests (<a href="https://togithub.com/onsi/ginkgo/issues/752">#752</a>) [efb9e69]
• Add a Chinese Doc <a href="https://togithub.com/onsi/ginkgo/issues/755">#755</a> (<a href="https://togithub.com/onsi/ginkgo/issues/756">#756</a>) [5207632]
• cli: allow multiple -focus and -skip flags (<a href="https://togithub.com/onsi/ginkgo/issues/736">#736</a>) [9a782fb]</p>
<p><strong>Fixes</strong></p>
<p>• Add _internal to filename of tests created with internal flag (<a href="https://togithub.com/onsi/ginkgo/issues/751">#751</a>) [43c12da]</p>
<p><strong><a href="https://togithub.com/onsi/ginkgo/releases/v1.14.2"><code>v1.14.2</code></a></strong></p>
<p><a href="https://togithub.com/onsi/ginkgo/compare/v1.14.1...v1.14.2">Compare Source</a></p>
<p><strong>Fixes</strong></p>
<p>• correct handling windows backslash in import path (<a href="https://togithub.com/onsi/ginkgo/issues/721">#721</a>) [97f3d51]
• Add additional methods to GinkgoT() to improve compatibility with the testing.TB interface [b5fe44d]</p>
<p><strong><a href="https://togithub.com/onsi/ginkgo/releases/v1.14.1"><code>v1.14.1</code></a></strong></p>
<p><a href="https://togithub.com/onsi/ginkgo/compare/v1.14.0...v1.14.1">Compare Source</a></p>
<p><strong>Fixes</strong></p>
<p>• Discard exported method declaration when running ginkgo bootstrap (<a href="https://togithub.com/onsi/ginkgo/issues/558">#558</a>) [f4b0240]</p>
<p><strong><a href="https://togithub.com/onsi/ginkgo/releases/v1.14.0"><code>v1.14.0</code></a></strong></p>
<p><a href="https://togithub.com/onsi/ginkgo/compare/v1.13.0...v1.14.0">Compare Source</a></p>
<p><strong>Features</strong></p>
<p>• Defer running top-level container nodes until RunSpecs is called [d44dedf]
• <a href="http://onsi.github.io/ginkgo/#understanding-ginkgos-lifecycle">Document Ginkgo lifecycle</a>
• Add <code>extensions/globals</code> package (<a href="https://togithub.com/onsi/ginkgo/issues/692">#692</a>) [3295c8f] - this can be helpful in contexts where you are test-driving your test-generation code (see <a href="https://togithub.com/onsi/ginkgo/pull/692">#692</a>)
• Print Skip reason in JUnit reporter if one was provided [820dfab]</p>
<p><strong><a href="https://togithub.com/onsi/ginkgo/releases/v1.13.0"><code>v1.13.0</code></a></strong></p>
<p><a href="https://togithub.com/onsi/ginkgo/compare/v1.12.3...v1.13.0">Compare Source</a></p>
<p><strong>Features</strong></p>
<p>• Add a version of table.Entry that allows dumping the entry parameters. (<a href="https://togithub.com/onsi/ginkgo/issues/689">#689</a>) [21eaef2]</p>
<p><strong>Fixes</strong></p>
<p>• Ensure integration tests pass in an environment sans GOPATH [606fba2]
• Add books package (<a href="https://togithub.com/onsi/ginkgo/issues/568">#568</a>) [fc0e44e]
• doc(readme): installation via "tools package" (<a href="https://togithub.com/onsi/ginkgo/issues/677">#677</a>) [83bb20e]
• Solve the undefined: unix.Dup2 compile error on mips64le (<a href="https://togithub.com/onsi/ginkgo/issues/680">#680</a>) [0624f75]
• Import package without dot (<a href="https://togithub.com/onsi/ginkgo/issues/687">#687</a>) [6321024]
• Fix integration tests to stop require GOPATH (<a href="https://togithub.com/onsi/ginkgo/issues/686">#686</a>) [a912ec5]</p>
<p><strong><a href="https://togithub.com/onsi/ginkgo/releases/v1.12.3"><code>v1.12.3</code></a></strong></p>
<p><a href="https://togithub.com/onsi/ginkgo/compare/v1.12.2...v1.12.3">Compare Source</a></p>
<p><strong>Fixes</strong></p>
<p>• Print correct code location of failing table test (<a href="https://togithub.com/onsi/ginkgo/issues/666">#666</a>) [c6d7afb]</p>
<p><strong><a href="https://togithub.com/onsi/ginkgo/releases/v1.12.2"><code>v1.12.2</code></a></strong></p>
<p><a href="https://togithub.com/onsi/ginkgo/compare/v1.12.1...v1.12.2">Compare Source</a></p>
<p><strong>Fixes</strong></p>
<p>• Update dependencies [ea4a036]</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/cucumber">here</a>.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-04 12:30:56
[cucumber/cucumber] Pull request ready for review by aslakhellesoy
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-04 12:34:10
[cucumber/cucumber] Comment on #1209
This is now ready for review!
The `@cucumber/react` code still needs a bit of work before this can be merged - I'm interested in a review of the other parts though!
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-05 02:33:15
[cucumber/cucumber] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-05 02:37:09
[cucumber/cucumber] Comment on #1499
Hey <a href="https://github.com/hkosova">@hkosova</a> 🙂
Thanks again for your contribution 👍
Do you think you could quickly update the changelog please?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-05 02:39:06
[cucumber/cucumber] Comment on #1513
Hello <a href="https://github.com/OrhanTozan">@OrhanTozan</a>
Thank you for your contribution.
Updating translation actually requires a little bit more changes.
Could you please take a look on other places where updates are required?
You can take some inspiration from another PR like the following one if you need to: <a href="https://github.com/cucumber/cucumber/pull/1499">#1499</a>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-05 02:48:20
[cucumber/cucumber] Comment on #1499
<a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> done
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-05 04:43:59
[cucumber/cucumber] Comment on #1209
In order to finish this we have to wait for <a href="https://github.com/remarkjs/react-markdown/pull/588">remarkjs/react-markdown#588</a> to be merged and released.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-05 05:27:50
[cucumber/cucumber] Comment on #1499
Thank you!
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-05 05:27:59
[cucumber/cucumber] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-05 05:28:00
[cucumber/cucumber] Comment on #1499
Hi <a href="https://github.com/hkosova">@hkosova</a>,
Thanks for your making your first contribution to Cucumber, and welcome to the Cucumber committers team! You can now push directly to this repo and all other repos under the cucumber organization! 🍾
In return for this generous offer we hope you will:
• ✅ Continue to use branches and pull requests. When someone on the core team approves a pull request (yours or someone else's), you're welcome to merge it yourself.
• 💚 Commit to setting a good example by following and upholding our <a href="https://cucumber.io/conduct">code of conduct</a> in your interactions with other collaborators and users.
• 💬 Join the <a href="http://cucumber.io/support#slack">community Slack channel</a> to meet the rest of the team and make yourself at home.
• ℹ️ Don't feel obliged to help, just do what you can if you have the time and the energy.
• 🙋 Ask if you need anything. We're looking for feedback about how to make the project more welcoming, so please tell us!
On behalf of the Cucumber core team,
Aslak Hellesøy
Creator of Cucumber
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-05 06:43:05
[cucumber/cucumber] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-05 12:40:30
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
onsi/gomega
<strong><a href="https://togithub.com/onsi/gomega/releases/v1.12.0"><code>v1.12.0</code></a></strong></p>
<p><a href="https://togithub.com/onsi/gomega/compare/v1.11.0...v1.12.0">Compare Source</a></p>
<p><strong>Features</strong></p>
<p>• Add Satisfy() matcher (<a href="https://togithub.com/onsi/gomega/issues/437">#437</a>) [c548f31]
• tweak truncation message [3360b8c]
• Add format.GomegaStringer (<a href="https://togithub.com/onsi/gomega/issues/427">#427</a>) [cc80b6f]
• Add Clear() method to gbytes.Buffer [c3c0920]</p>
<p><strong>Fixes</strong></p>
<p>• Fix error message in BeNumericallyMatcher (<a href="https://togithub.com/onsi/gomega/issues/432">#432</a>) [09c074a]
• Bump <a href="http://github.com/onsi/ginkgo">github.com/onsi/ginkgo</a> from 1.12.1 to 1.16.2 (<a href="https://togithub.com/onsi/gomega/issues/442">#442</a>) [e5f6ea0]
• Bump <a href="http://github.com/golang/protobuf">github.com/golang/protobuf</a> from 1.4.3 to 1.5.2 (<a href="https://togithub.com/onsi/gomega/issues/431">#431</a>) [adae3bf]
• Bump <a href="http://golang.org/x/net">golang.org/x/net</a> (<a href="https://togithub.com/onsi/gomega/issues/441">#441</a>) [3275b35]</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/cucumber">here</a>.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-05 16:24:41
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/cucumber">here</a>.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-05 21:30:00
[cucumber/cucumber] Comment on #1507
<a href="https://github.com/mattwynne">@mattwynne</a> I've gone through godog to update the branch and text references to main but it appears I may not have the permissions needed to actually update the repo on the GH side anymore. GH wants me to try making a PR. 😅
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 00:25:48
[cucumber/cucumber] SabotageAndi approved gasparnagy's pull request
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 00:32:46
[cucumber/cucumber] Issue opened by Yellaks
<p>The latest 'messages-go' release (v15.0.0) has a go.mod dependancy on Protobuf v1.3.1 which has a security issue, flagged here: <a href="https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-3121">https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-3121</a></p>
<p>The master branch does not include this dependancy and has been removed in this commit: <a href="https://github.com/cucumber/messages-go/commit/9586c3b101a18dfe21b0d04735759be69b35210f">cucumber/messages-go@9586c3b</a></p>
<p>When is a new/next release expected?</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 00:42:36
[cucumber/cucumber] Comment on #1505
Normally Gherkin allows you to make full line comments only.
Commenting after tags on the same line was specially hacked in because many IDEs allowed this and it caused confusion. But since the current Gherkin parser uses the lines as tokens, it is not easy to handle these "special" comments as normal ones. So I'm afraid there is no easy fix for that. I encourage everyone to NOT use this kind of commenting... I recommend using the `@tags_with_spaces style` instead.
Maybe <a href="https://github.com/mpkorstanje">@mpkorstanje</a> can add something to this.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 00:42:45
[cucumber/cucumber] Comment on #1413
Yes, this seem to be disallowed by the grammar. I don't know whether this was intentional though.
<a href="https://github.com/cucumber/cucumber/blob/master/gherkin/gherkin.berp#L37">https://github.com/cucumber/cucumber/blob/master/gherkin/gherkin.berp#L37</a>
```
FeatureHeader! := #Language? Tags? <b>#FeatureLine</b> DescriptionHelper
...
DescriptionHelper := #Empty* Description? #Comment*
```
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 01:05:04
[cucumber/cucumber] Pull request ready for review by gasparnagy
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 01:06:58
[cucumber/cucumber] Comment on #1513
<a href="https://github.com/OrhanTozan">@OrhanTozan</a> Check <a href="https://github.com/cucumber/cucumber/blob/master/gherkin/CONTRIBUTING.md#adding-or-updating-an-i18n-language">https://github.com/cucumber/cucumber/blob/master/gherkin/CONTRIBUTING.md#adding-or-updating-an-i18n-language</a> that describes the steps to be done to distribute the change to all languages
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 01:43:19
[cucumber/cucumber] Issue opened by gasparnagy
<p>I want to share my thoughts on this topic, as a kind of discussion forum here.</p>
<p><strong>Defaults in Messages JSON</strong></p>
<p>As I worked a bit with the new messages structure, I would like to bring up (again) the problem of defaults in the messages protocol. Currently I have found 3 cases that were not fully consistent with my understanding:</p>
<p>• Scenario[/Background/...].description
• Scenario[/Background/...].name
• TableCell.value</p>
<p>These are currently set as required in the JSON schema, although they are optional in Gherkin and in many cases they are usually not provided (e.g. I have never seen anyone adding a name or description to a Background, but also Scenario description is rarely used). Forcing ourselves to include a dummy empty setting for them in the messages JSON files seems to be unnecessary for me.</p>
<p>As far as I have understood <a href="https://github.com/aslakhellesoy">@aslakhellesoy</a>, the original intention was to populate these properties for dynamic languages, but with TS this does not seem to be necessary anyway.</p>
<p>A similar question could be raised for the empty arrays, but those are more correct semantically.</p>
<p>Location.column is currently set as optional (inconsistency with the others)?</p>
<p>Also I feel that if we make "set everything as required" as a rule for the messages JSON, then if later we introduce a new (optional) setting (and make it required by following our rule) then the old JSON message files will became invalid according to the schema, so it will be harder to introduce non-breaking changes.</p>
<p>So my questions are:</p>
<p>• Shall we make these mentioned string fields optional in the JSON schema?
• Shall we make the arrays optional in the JSON schema?</p>
<p><strong>Defaults in Messages objects</strong></p>
<p>If we would make these settings as optional in the JSON schema, we could still think about the default value that we use in the populated message objects if the particular field was not specified. As far as I know JSON schema does not provide an option to explicitly specify the default value, so this can only be an agreement that the different implementations follow.</p>
<p>I think here there are 3 questions:</p>
<p>• What should we use as default value for not-provided optional <strong>strings</strong> (empty string or null)?
• What should we use as default value for nor-provided optional <strong>arrays</strong> (empty array or null)?
• What should we use as default value for nor-provided optional <strong>numbers</strong> (0 or null -- for null we need to represent it as nullable number in the objects for static type languages)?</p>
<p>What do you think?</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 01:53:45
[cucumber/cucumber] Comment on #1519
Good question. So my thoughts / observations.
100% agree Description on Background - I've never once seen it in 5 years
Scenario description "can" be used. And I've seen it occassionally, but then it's not always picked up nicely by html formatters e.t.c. (Not tested on new one).
IMO almost any string field in Gherkin should be optional, save for things which come after Gherkin keywords.
Answers for default
• Empty string
• Empty array
• .... Tricky one. I would probably lean more towards `null` but I'm not the best person to ask about how much extra work this entails. Ruby (primary language), is quite flexible in this regards. I appreciate static ones might need a lot more skeleton work for this.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 02:16:45
[cucumber/cucumber] Issue opened by punkratz312
<p>please add the possibility to define hard coded execution oder as last workaround. its not hard to implement and wont break general approach that tests need to be independent 💯 right.</p>
<p>As always there might be an exception in case of an initial password change its mandatory that the update password scenario runs first. As otherwise all the other tests would need to deal with this situation too and thats not practicable.</p>
<p>Please implement the feature options they are interpreted all over the internet even tho it has not any effect at all, to define multiple feature blocks. Still the ugly number_prefix workaround must be used to workaround this special case...</p>
<p>Desired solution:</p>
<p><a href="https://user-images.githubusercontent.com/8830888/117272674-b3fea880-ae5b-11eb-92c6-6d2e0b0ad54b.png">image</a></p>
<p>Actual workaround:</p>
<p><a href="https://user-images.githubusercontent.com/8830888/117273272-469f4780-ae5c-11eb-8bd2-9c71bbf8b33f.png">image</a></p>
<p><a href="https://forum.katalon.com/t/how-to-run-multiple-feature-files-in-a-specific-sequence-using-cucumber-runner-class/24327">https://forum.katalon.com/t/how-to-run-multiple-feature-files-in-a-specific-sequence-using-cucumber-runner-class/24327</a></p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 02:20:22
[cucumber/cucumber] Comment on #1505
I agree these comments don't adhere to the grammar. Unfortunately, I have limited influence how the files are written.
As a workaround, we could have a flag that _disables_ the hack that allows such comments.
Then they would be reported as parsing errors, and we at least know we're losing some content.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 04:48:11
[cucumber/cucumber] Comment on #1505
As this is deep in the grammar level, the usual cucumber settings cannot be applied there... so it is not easy to make a setting for that (the grammar/parser can be also used even without cucumber)
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 06:53:59
[cucumber/cucumber] Comment on #1505
As trade-offs go I think we currently have the right one. The user experience should weight much higher then being able to build a a feature file from the AST.
Though I do see some room in the current AST to allow the comments on tag lines to be presented included. The `location` element uses both lines and columns. So perhaps the tag comments can be included as comments starting at a non-zero column?
```
/**
* A comment in a Gherkin document
*/
message Comment {
// The location of the comment
Location location = 1;
// The text of the comment
string text = 2;
}
/**
* Points to a line and a column in a text file
*/
message Location {
uint32 line = 1;
uint32 column = 2;
}
```
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 07:54:10
[cucumber/cucumber] Issue opened by aurelien-reeves
<p>Our README and CONTRIBUTING documents may certainly be more welcome, and standardize.</p>
<p>Would anyone have ideas? Or source for inspiration?</p>
<p>My self I've spotted how Nextcloud is welcoming contributors with a dedicated landing page, and a developer manual</p>
<p>• <a href="https://nextcloud.com/contribute/">https://nextcloud.com/contribute/</a>
• <a href="https://docs.nextcloud.com/server/latest/developer_manual/getting_started/index.html">https://docs.nextcloud.com/server/latest/developer<em>manual/getting</em>started/index.html</a></p>
<p>However that may request some maintenance on a regular basis</p>
<p>And Ruby on Rails also has a dedicated page on their website with a lot of technical details for incoming developers</p>
<p>• <a href="https://edgeguides.rubyonrails.org/contributing_to_ruby_on_rails.html">https://edgeguides.rubyonrails.org/contributing<em>to</em>ruby<em>on</em>rails.html</a></p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 07:57:06
[cucumber/cucumber] Comment on #1519
Broadly agree with <a href="https://github.com/gasparnagy">@gasparnagy</a> regarding defaults.
Specifically my feeling would be:
• string - don't default to empty string
• array - don't feel strongly either way
> Location.column is currently set as optional (inconsistency with the others)?
I think there was a functional reason for this that <a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> mentioned the other week but I don't remember what?
With descriptions in Scenario, Background et al, I think at runtime for building reports we would be checking whether there is a non-blank string (i.e. has content other than whitespace) so I don't think we'd be "saving" a check there by defaulting to an empty string.
TypeScript-specific aside: if we were to drop the idea of defaults on deserialisation, the TypeScript classes generated from the JSON schema could just become interfaces, and we'd no longer have the weight of those classes in the compiled JavaScript (though this saving is probably an order of magnitude less than the one from losing the protobufjs runtime).
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 08:08:26
[cucumber/cucumber] Comment on #1521
On the README side of things for users, I think the cucumber-js one could really use an overhaul.
I really like the README for ava <a href="https://github.com/avajs/ava">https://github.com/avajs/ava</a>
In particular:
• Logo reassures you found the right thing
• No build/stats badges - [whispers] I tend to view them as just clutter these days
• Little animation showing the tool in action
• Enough info to get you started with a test that runs, and then links to detailed docs
• Early callout to potential contributors to funnel them into the CONTRIBUTING doc
WDYT?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 08:10:53
[cucumber/cucumber] Comment on #1521
I agree.
We are focus on contributors, but I still think READMEs should introduce properly the tool, quickly, to users before redirecting them to the documentation website.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 08:37:29
[cucumber/cucumber] aslakhellesoy approved gasparnagy's pull request
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 08:37:42
[cucumber/cucumber] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 08:40:08
[cucumber/cucumber] Comment on #1391
Neat trick!
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 08:42:41
[cucumber/cucumber] Comment on #1391
Love this! I assume this makes <a href="https://github.com/cucumber/cucumber/pull/1179">#1179</a> obsolete?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 08:43:47
[cucumber/cucumber] Comment on #1391
I think we should add `cpy-cli` to `devDependencies` to avoid the prompt if we want to install it?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 08:45:55
[cucumber/cucumber] Comment on #1391
should this be `docString`
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 08:56:41
[cucumber/cucumber] Comment on #1507
<a href="https://github.com/jaysonesmith">@jaysonesmith</a> so you mean it wouldn't let you change the default branch? Or something else?
I've changed your org role from Member to Owner, so I think you should be good now. Please use your new powers responsibly :)
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 09:10:16
[cucumber/cucumber] Comment on #1391
Just the CSS classname but yes it should be consistent, will change.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 09:10:54
[cucumber/cucumber] Comment on #1391
Yep, I think so.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 09:12:16
[cucumber/cucumber] Comment on #1521
Great - I'll run up a PR for cucumber-js soon and will try and come back with some thoughts on contrubuting docs too.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 11:12:20
[cucumber/cucumber] Issue opened by mattwynne
<p>As part of our strategy to improve the experience for new contributors, I propose we rename this repo to better reflect our content intentions for it, which is to contain the shared or common components that are used by the plaform-specific cucumber flavours.</p>
<p>I have no idea what trouble will be caused in terms of failing builds etc bu this change, but I think there's only one way to find out. If it proves too difficult we can just roll back :)</p>
<p>I suggest <code>common</code> as the new name. Any concerns / objections or better ideas?</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 11:26:07
[cucumber/cucumber] Issue opened by mattwynne
<p>We have some really ancient PRs and Issues in some repos, e.g. <a href="https://github.com/cucumber/cucumber-cpp/pull/129">cucumber/cucumber-cpp#129</a></p>
<p>I'd like to use Stalebot to help us manage this, and I think it would be good to try and have a consistent rule set for all our repos.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 12:13:09
[cucumber/cucumber] Comment on #1507
I discovered that renaming the branch using GitHub's UI (in /settings/branches) is probably the way to go. It automatically updates any PRs against the old branch, and displays a message to users letting them what to do to update their settings.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 12:13:14
[cucumber/cucumber] Comment on #1505
<a href="https://github.com/mpkorstanje">@mpkorstanje</a> I think the AST would be able to handle the comments (BTW the normal comments do not necessarily start at column 0 or 1, but the column value indicates the position of the `#` sign within the line), the more problem is how the token scanner can feed the comment token back to the parser... Anyway. I totally agree with you that we have the right trade-off.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 12:15:32
[cucumber/cucumber] Comment on #1427
<a href="https://github.com/declark">@declark</a> has shared <a href="https://cucumberbdd.slack.com/archives/C62D0FK0E/p1619625179280100">a video in Slack</a> about their proposed changes to improve the UX of the SmartBear community site:
<a href="https://smartbear.wistia.com/medias/h9wmm5w1wb">https://smartbear.wistia.com/medias/h9wmm5w1wb</a>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 12:27:52
[cucumber/cucumber] Comment on #1507
godog is done! 🎉
<a href="https://user-images.githubusercontent.com/21065409/117354644-6dfb1200-ae66-11eb-94c9-f8a9a70543e1.png">Screen Shot 2021-05-06 at 12 27 17 PM</a>
Thanks for the role update <a href="https://github.com/mattwynne">@mattwynne</a>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 12:29:31
[cucumber/cucumber] Comment on #1519
Personally I have an aversion to nulls in general: they breed `if` statements, which in turn breed bugs. But in a protocol like this there will be some places they're appropriate - the backwards compatibility case you've mentioned <a href="https://github.com/gasparnagy">@gasparnagy</a> being one of those.
What's important to me is the semantics of the protocol - an empty string or array and a null mean different things, and we should be clear about which meaning we _want_ to give. So I think we might need to look at each particular field in the messages where we might need to handle nulls, not just string/array/numbers in general.
From the semantics perspective. I don't see the problem with representing "the user did not specify a description for this scenario" with an empty string. It's not like the user could tell us, in their Gherkin document, _I want to have a description for this scenario, and I want it to be blank_ (i.e. empty string) versus _I don't want this scenario to have a description_ (i.e. null). Either they give us a description or they don't.
Similarly empty arrays seem like a fine way to represent nothingness in all the cases I can think of - e.g. the tags for a scenario.
Numbers, I think are different. Zero is a fairly arbitrary default value, and implies more knoweledge of the answer than might be appropriate. A null value for `location.column` says "I don't know (or cannot say for this language) what column the thing is located at" wheras a zero imples "I know where it's located, it's located in the first column".
Them's my thoughts, does that help?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 12:33:39
[cucumber/cucumber] Comment on #1522
Metaphorically, I guess "cells" would be a nice name - A cucumber is made of cells :)
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 12:43:46
[cucumber/cucumber] Comment on #1507
<a href="https://github.com/luke-hill">@luke-hill</a> could you take care of cucumber-rails?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 19:20:06
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
protocolbuffers/protobuf
<strong><a href="https://togithub.com/protocolbuffers/protobuf/releases/v3.16.0"><code>v3.16.0</code></a></strong></p>
<p><a href="https://togithub.com/protocolbuffers/protobuf/compare/v3.15.8...v3.16.0">Compare Source</a></p>
<p><strong>C++</strong></p>
<p>• Fix compiler warnings issue found in conformance<em>test</em>runner <a href="https://togithub.com/protocolbuffers/protobuf/issues/8189">#8189</a> (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8190">#8190</a>)
• Fix MinGW-w64 build issues. (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8286">#8286</a>)
• [Protoc] C++ Resolved an issue where NO<em>DESTROY and CONSTINIT are in incorrect order (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8296">#8296</a>)
• Fix PROTOBUF</em>CONSTINIT macro redefinition (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8323">#8323</a>)
• Delete StringPiecePod (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8353">#8353</a>)
• Fix gcc error: comparison of unsigned expression in '>= 0' is always … (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8309">#8309</a>)
• Fix cmake install on iOS (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8301">#8301</a>)
• Create a CMake option to control whether or not RTTI is enabled (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8347">#8347</a>)
• Fix endian.h location on FreeBSD (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8351">#8351</a>)
• Refactor util::Status (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8354">#8354</a>)
• Make util::Status more similar to absl::Status (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8405">#8405</a>)
• Fix -Wsuggest-destructor-override for generated C++ proto classes. (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8408">#8408</a>)
• Refactor StatusOr and StringPiece (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8406">#8406</a>)
• Refactor uint128 (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8416">#8416</a>)
• The ::pb namespace is no longer exposed due to conflicts.
• Allow MessageDifferencer::TreatAsSet() (and friends) to override previous<br />
calls instead of crashing.
• Reduce the size of generated proto headers for protos with <code>string</code> or<br />
<code>bytes</code> fields.
• Move arena() operation on uncommon path to out-of-line routine
• For iterator-pair function parameter types, take both iterators by value.
• Code-space savings and perhaps some modest performance improvements in<br />
RepeatedPtrField.
• Eliminate nullptr check from every tag parse.
• Remove unused <em>$name$</em>cached<em>byte</em>size_ fields.
• Serialize extension ranges together when not broken by a proto field in the<br />
middle.
• Do out-of-line allocation and deallocation of string object in ArenaString.
• Streamline ParseContext::ParseMessage to avoid code bloat and improve<br />
performance.
• New member functions RepeatedField::Assign, RepeatedPtrField::{Add, Assign}.
• Fix undefined behavior warning due to innocuous uninitialization of value<br />
on an error path.
• Avoid expensive inlined code space for encoding message length for messages</p>
<pre><code>&gt; = 128 bytes and instead do a procedure call to a shared out-of-line routine.
</code></pre>
<p>• util::DefaultFieldComparator will be final in a future version of protobuf.<br />
Subclasses should inherit from SimpleFieldComparator instead.</p>
<p><strong>C</strong></p>
<p>• Add .NET 5 target and improve WriteString performance with SIMD (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8147">#8147</a>)</p>
<p><strong>Java</strong></p>
<p>• deps: update JUnit and Truth (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8319">#8319</a>)
• Detect invalid overflow of byteLimit and return InvalidProtocolBufferException as documented.
• Exceptions thrown while reading from an InputStream in parseFrom are now<br />
included as causes.
• Support potentially more efficient proto parsing from RopeByteStrings.
• Clarify runtime of ByteString.Output.toStringBuffer().
• Added UnsafeByteOperations to protobuf-lite (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8426">#8426</a>)</p>
<p><strong>JavaScript</strong></p>
<p>• Make Any.pack() chainable.</p>
<p><strong>Python</strong></p>
<p>• Fix some constness / char literal issues being found by MSVC standard conforming mode (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8344">#8344</a>)
• Switch on "new" buffer API (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8339">#8339</a>)
• Enable crosscompiling aarch64 python wheels under dockcross manylinux docker image (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8280">#8280</a>)
• Fixed a bug in text format where a trailing colon was printed for repeated field.
• When TextFormat encounters a duplicate message map key, replace the current<br />
one instead of merging.</p>
<p><strong>Objective-C</strong></p>
<p>• Move the class map to a CFDictionary. (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8328">#8328</a>)</p>
<p><strong>PHP</strong></p>
<p>• read_property() handler is not supposed to return NULL (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8362">#8362</a>)
• Changed parameter type from long to integer (<a href="https://togithub.com/protocolbuffers/protobuf/issues/7613">#7613</a>)
• fix: README supported PHP version for C extension (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8236">#8236</a>)</p>
<p><strong>Ruby</strong></p>
<p>• Fixed quadratic memory usage when appending to arrays. (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8364">#8364</a>)
• Fixed memory leak of Ruby arena objects. (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8461">#8461</a>)
• Add support for proto3 json_name in compiler and field definitions. (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8356">#8356</a>)</p>
<p><strong>Other</strong></p>
<p>• Some doc on AOT compilation and protobuf (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8294">#8294</a>)
• [CMake] Ability to pass options to protoc executable from cmake (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8374">#8374</a>)
• Add --fatal_warnings flag to treat warnings as errors (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8131">#8131</a>)
• [bazel] Remove deprecated way to depend on googletest (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8396">#8396</a>)
• add error returns missing from protoc to prevent it from exiting with… (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8409">#8409</a>)</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/cucumber">here</a>.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-06 23:54:36
[cucumber/cucumber] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 03:06:45
[cucumber/cucumber] Comment on #1507
Think I have same issue as Jayson. Not got the 🔋
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 03:42:43
[cucumber/cucumber] Comment on #1507
<a href="https://github.com/mattwynne">@mattwynne</a> I lack the permissions on cucumber-js as well.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 05:42:09
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about these updates again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/cucumber">here</a>.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 05:42:20
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/cucumber">here</a>.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 05:42:44
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about these updates again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/cucumber">here</a>.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 05:42:55
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
nodejs/node
<strong><a href="https://togithub.com/nodejs/node/blob/master/CHANGELOG.md#%E2%80%8B2016-05-06-Version-01214-Maintenance--rvagg|<code">https://togithub.com/nodejs/node/blob/master/CHANGELOG.md#%E2%80%8B2016-05-06-Version-01214-Maintenance--rvagg|<code</a>v16</code>></strong></p>
<p><doc/changelogs/CHANGELOG<em>V012.md#0.12.14|Moved to doc/changelogs/CHANGELOG</em>V012.md#0.12.14>.</p>
<p><strong><a href="https://togithub.com/nodejs/node/blob/master/CHANGELOG.md#%E2%80%8B2015-12-23-Version-424-Argon-LTS--jasnell|<code">https://togithub.com/nodejs/node/blob/master/CHANGELOG.md#%E2%80%8B2015-12-23-Version-424-Argon-LTS--jasnell|<code</a>v15</code>></strong></p>
<p><doc/changelogs/CHANGELOG<em>V4.md#4.2.4|Moved to doc/changelogs/CHANGELOG</em>V4.md#4.2.4>.</p>
<p><strong><a href="https://togithub.com/nodejs/node/blob/master/CHANGELOG.md#%E2%80%8B2016-05-06-Version-01214-Maintenance--rvagg|<code">https://togithub.com/nodejs/node/blob/master/CHANGELOG.md#%E2%80%8B2016-05-06-Version-01214-Maintenance--rvagg|<code</a>v14</code>></strong></p>
<p><doc/changelogs/CHANGELOG<em>V012.md#0.12.14|Moved to doc/changelogs/CHANGELOG</em>V012.md#0.12.14>.</p>
<p><strong><a href="https://togithub.com/nodejs/node/blob/master/CHANGELOG.md#%E2%80%8B2016-03-31-Version-01213-LTS--rvagg|<code">https://togithub.com/nodejs/node/blob/master/CHANGELOG.md#%E2%80%8B2016-03-31-Version-01213-LTS--rvagg|<code</a>v13</code>></strong></p>
<p><doc/changelogs/CHANGELOG<em>V012.md#0.12.13|Moved to doc/changelogs/CHANGELOG</em>V012.md#0.12.13>.</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/cucumber">here</a>.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 05:43:06
[cucumber/cucumber] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about these updates again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/cucumber">here</a>.</p>
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 05:43:22
[cucumber/cucumber] Pull request closed by renovate[bot]
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 06:10:03
[cucumber/cucumber] Comment on #1521
The <a href="https://github.com/atom/atom/blob/master/CONTRIBUTING.md">contributing guide for atom</a> is pretty great!
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 06:59:35
[cucumber/cucumber] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 06:59:56
[cucumber/cucumber] Pull request closed by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 07:01:11
[cucumber/cucumber] Pull request closed by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 07:01:33
[cucumber/cucumber] Comment on #1527
*Renovate Ignore Notification*
As this PR has been closed unmerged, Renovate will ignore this upgrade and you will not receive PRs for _any_ future 15.x releases. However, if you upgrade to 15.x manually then Renovate will then reenable updates for minor and patch updates automatically.
If this PR was closed by mistake or you changed your mind, you can simply rename this PR and you will soon get a fresh replacement PR opened.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 07:01:33
[cucumber/cucumber] Pull request closed by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 07:02:22
[cucumber/cucumber] Comment on #1528
Should we give it a try to that one?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 07:02:53
[cucumber/cucumber] Comment on #1525
*Renovate Ignore Notification*
As this PR has been closed unmerged, Renovate will now ignore this update (2.7). You will still receive a PR once a newer version is released, so if you wish to permanently ignore this dependency, please add it to the `ignoreDeps` array of your renovate config.
If this PR was closed by mistake or you changed your mind, you can simply rename this PR and you will soon get a fresh replacement PR opened.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 07:03:01
[cucumber/cucumber] Comment on #1529
*Renovate Ignore Notification*
As this PR has been closed unmerged, Renovate will ignore this upgrade and you will not receive PRs for _any_ future 3.x releases. However, if you upgrade to 3.x manually then Renovate will then reenable updates for minor and patch updates automatically.
If this PR was closed by mistake or you changed your mind, you can simply rename this PR and you will soon get a fresh replacement PR opened.
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 08:04:31
[cucumber/cucumber] aslakhellesoy commented on their pull request
Reviewed with <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> and <a href="https://github.com/mattwynne">@mattwynne</a>
• Call it "Markdown with Gherkin" (MdG)
• Rename token matcher
• Document why we're not matching dialect (has to be set globally by cucumber, see cucumber-js has a command line option for this).
• Document that JSON formatters won't have descriptions for Markdown documents.
• Tell people who want it that JSON formatter is in maintenance mode - use message format instead.
• Maybe add descriptions to AST, it might be easy...
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 08:05:34
[cucumber/cucumber] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 10:08:15
[cucumber/cucumber] Comment on #1511
Great to see progress on this <a href="https://github.com/gasparnagy">@gasparnagy</a>!
Do you plan to migrate away from protobuf and use code generation from the JSON Schemas in this PR? If not, do you plan to do this in a different PR?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 10:12:32
[cucumber/cucumber] Comment on #1507
OK <a href="https://github.com/luke-hill">@luke-hill</a> <a href="https://github.com/davidjgoss">@davidjgoss</a> I think neither of you had been added to the <a href="https://github.com/orgs/cucumber/teams/core">@cucumber/core</a> team in github. I've fixed that - can you try again now?
<a href="https://github.com/cucumber/cucumber">cucumber/cucumber</a>
GitHub
2021-05-07 10:14:18
[cucumber/common] Comment on #1522
OK, I've renamed it to `common` - let's see how that sits.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-07 10:14:24
[cucumber/common] Issue closed by mattwynne
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-07 10:22:14
[cucumber/common] Comment on #1507
cucumber-js done!
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-07 10:52:36
[cucumber/common] Comment on #1511
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> that will be another PR. in this one i did not touch messages. .NET Gherkin did not have a reference to the messages so far, so I could do this fixes/cleanup without fixing messages first. But the structures that the improved Gherkin uses are essentially the same as we will have in messages generated from the schema (I manually copied the structure of the java generated classes for now).
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-07 13:01:31
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-08 10:00:08
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
sass/node-sass
<strong><a href="https://togithub.com/sass/node-sass/releases/v6.0.0"><code>v6.0.0</code></a></strong></p>
<p><a href="https://togithub.com/sass/node-sass/compare/v5.0.0...v6.0.0">Compare Source</a></p>
<p><strong>Breaking changes</strong></p>
<p>• Drop support for Node 10 (<a href="https://togithub.com/nschonni">@nschonni</a>)
• Remove deprecated process.sass API (<a href="https://togithub.com/xzyfer">@xzyfer</a>, <a href="https://togithub.com/sass/node-sass/issues/2986">#2986</a>)</p>
<p><strong>Features</strong></p>
<p>• Add support for Node 16</p>
<p><strong>Community</strong></p>
<p>• Fix typos in Troubleshooting guide (<a href="https://togithub.com/independencyinjection">@independencyinjection</a>, <a href="https://togithub.com/sass/node-sass/issues/3051">#3051</a>)
• Improve dependabot configuration (<a href="https://togithub.com/nschonni">@nschonni</a>)</p>
<p><strong>Supported Environments</strong></p>
<p>**Linux support refers to major distributions like Ubuntu, and Debian</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-09 08:07:35
[cucumber/common] Comment on #1520
> please add the possibility to define hard coded execution oder as last workaround. its not hard to implement and wont break general approach that tests need to be independent 💯 right.
Please submit a pull request if you want to see this added to Cucumber.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-09 13:21:51
[cucumber/common] Comment on #1519
It looks to me that JSON Schema supports default values. See <a href="https://json-schema.org/draft/2020-12/json-schema-validation.html#rfc.section.9.2">https://json-schema.org/draft/2020-12/json-schema-validation.html#rfc.section.9.2</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-09 22:18:22
[cucumber/common] Comment on #1510
Thanks for doing this <a href="https://github.com/davidjgoss">@davidjgoss</a>!
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-10 00:47:36
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-10 02:10:11
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-10 02:57:10
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-10 05:03:42
[cucumber/common] Comment on #1519
I think this is easier to reason about if we think about the various roles
our components play:
• Message producers
• Gherkin Parser (produces `source` and `gherkinDocument` messages)
• Pickle Compiler (produces `pickle` messages)
• Cucumber (produces all the other message types)
• In-Process message consumers
• Pickle Compiler (consumes `gherkinDocument` messages from the Gherkin Parser)
• Cucumber (consumes `pickle` messages)
• Out of process message consumers
• Cucumber-React (JavaScript)
• Standalone JSON Formatter (Go)
• Future components yet to be written
If we only had in-process message consumers, we wouldn't need to worry about required fields and default values. There wouldn't be any JSON serialisation or deserialisation. If the producer didn't provide a value for a property, it would either be caught by the compiler (TypeScript), or if the language doesn't prevent assigning null values to fields (Java, not sure about C#) it would likely be caught by a null reference in a test.
It's primarily the out-of process message consumers we need to worry about. They should assume that the messages they consume contain garbage.
*Default values*
We can reduce the amount of garbage with default values. When we parse a JSON string into an object, any property missing from the JSON would get the default value from the JSON Schema.
This is similar to how the protobuf deserialisation works. After deserialisation, any missing property gets the "null" value by default (`[]` for arrays, `0` for numbers and `""` for strings). The main difference is that the default value is not explicitly declared in the schema, but follows a convention defined by the protobuf standard.
We have a choice how we want to deal with default values. We can either mimic the protobuf behaviour (i.e. conventional default values), or we can rely on the `default` value from the schema.
I prefer the explicit style since it doesn't require programmers to know about any conventions. It also gives us more fine-grained control at the cost of being a little more verbose in the schema.
*Required properties*
Declaring a property as required in the schema offers no guarantees to the consumer that the property will have a value in the JSON. It could however give the consumer the confidence that the deserialised object has a value for all required properties.
If we want to avoid consumers littered with conditionals checking if a required property actually has a value after deserialising JSON, we can:
1. Provide a `default` value for `required` properties that were absent from the JSON.
2. Throw an error if a `required` property was absent _and_ the schema didn't specify a `default`.
With this in place, consumers can safely assume that required properties will always have a value after serialisation.
This means that the deserialisers need to know about the schema.
*What does this mean in practice?*
I think we should go over the schemas and apply some rules to decide whether a particular property should be `required`, have a `default` or both. Here is my proposal:
*`required`, no `default`*
Any property where the absence of a value would mean the consumer cannot do the job properly. Examples are `id`, `timestamp` and `duration` properties.
This cannot be used for properties added in future versions.
*`required`, with `default`*
Any property that's optional where we want to avoid a null check in the consumer. I think this goes for most of the properties, for the reasons <a href="https://github.com/mattwynne">@mattwynne</a> has described. The `default` value declared in the schema would typically be `[]`, `0` or `""` (protobuf style).
Examples are `name`, `title` and `description`
*not `required`, no `default`*
I think we should avoid this since it requires consumer conditionals.
*not `required`, with `default`*
This makes no sense and should be avoided.
*Future properties*
Whenever we add new properties to the schema, we have to make sure new consumers (with the new property) can consume messages from old producers.
We have the following options:
• With `default` - no need for consumer conditionals
• Without `default` - needs consumer conditionals
I think we should evaluate on a case by case basis which option is best
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-10 05:49:11
[cucumber/common] Comment on #1519
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> Thank you for the summary.
I generally agree with you, except one thing. You said, we anyway cannot "guarantee" that producers will include the required properties. I think this is not that simple. The schema is part of the specification. If we have a schema, that declares a property as "required" that means that by the specification of the cucumber messages format, the users MUST provide that value. Producers can decide to not provide the value of course, but that means their input will not conform the expectations for the message, so the result is somewhat unpredictable. (Showing a nice error message is of course nice.) If we publish the JSON schema than both the producers and the consumers can use it to (automatically) verify if the message fulfills that. This is standard for XML message validation, probably there are libraries in the different platforms where such check can be easily included in the deserialization process (ie. check if the JSON message conforms the schema and only start the deserialization if it does).
Based on that, I think we should declare in the schema what we expect and not what we might get. I.e. if a property is not required (e.g. description), than we should not mark it as "required" in the JSON schema, because with that we confuse the ones who want to use the schema as a contract (i.e. the message would fail the schema validation although the implementations would accept it). Because of that, I would NOT have anything as `required` with `default` but encourage consumers to reject any message that does not provide these values. So for `name`, `description` -> not `required` with `default` and do not use category `required` with `default`
I agree to the explicit mention of the default values (wherever possible), and I think we have a consensus on using empty string for string properties and empty array for lists as default. The only one I puzzle with is the `location.column`, where `0` is misleading (like <a href="https://github.com/luke-hill">@luke-hill</a> and <a href="https://github.com/mattwynne">@mattwynne</a> also mentioned). Technically `0` would not cause a trouble, because line position is "traditionally" specified as a 1-based index, so there is not 0 column, but it is misleading. So my vote here is to use `null`, but as this is a minor detail anyway, I am also fine with 0.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-10 09:38:33
[cucumber/common] Issue opened by mattwynne
<https://github.com/cucumber/common/issues/1532|#1532 Remove mono dependency from the build process>
<p>This is one of the pieces getting in the way of <a href="https://github.com/cucumber/cucumber-build/pull/12">cucumber/cucumber-build#12</a></p>
<p>Mono was historically there to let us run the Berp executable, but there may be other was to do it.</p>
<p><a href="https://github.com/gasparnagy">@gasparnagy</a> is going to have a look at whether we can use the newly-added .NET 5 runtime dependency to run Berp instead, or if not what our other options are.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-10 09:55:24
[cucumber/common] Comment on #1532
Yeah I don’t think we need mono anymore...
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-10 13:42:52
[cucumber/common] Issue opened by FrankieDcom
<p>Hello we have recently upgraded to Cucumber 7, and have started using the out of the box HTML formatter for our test reports. We also have retries enabled on our tests.</p>
<p>We are noticing when a test fails on the first run, but passes on a retry, the HTML formatter incorrectly records it as a failed test, while the summary recording it as a success. With a simple scenario:</p>
<pre><code>var count = 0;
When('Will fail first run', function () {
count++;
if (count &lt; 2) {
throw Error(`Fail on first try ${count}`);
}
});
</code></pre>
<pre><code>Feature: Test retry
Will fail on the first run
Scenario: Fail first run
When Will fail first run
</code></pre>
<p>Running the following generates the following report and summary:</p>
<p><a href="https://user-images.githubusercontent.com/36390015/108447007-005a3380-72c4-11eb-934e-716175afbc43.png">image</a></p>
<p><a href="https://user-images.githubusercontent.com/36390015/108447069-1831b780-72c4-11eb-85a9-394b2f37b426.png">image</a></p>
<p>As you can see, the HTML report displays a failure while the summary correctly shows the scenario as passing with a warning.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-11 00:50:37
[cucumber/common] Comment on #1533
> I just want to add that the same thing happens if we use builtin message formatter.
<a href="https://github.com/thiagojacinto">@thiagojacinto</a> could you please clarify what you mean by that? The `message` formatter outputs <a href="http://ndjson.org/">ndjson</a> messages. Are you seeing something incorrect in the outputted messages?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-11 03:05:54
[cucumber/common] Pull request ready for review by gasparnagy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-11 04:09:51
[cucumber/common] Comment on #1179
This is superseded by <a href="https://github.com/cucumber/common/pull/1391">#1391</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-11 04:09:51
[cucumber/common] Pull request closed by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-11 04:11:08
[cucumber/common] Comment on #1511
This looks good to me! Can we merge it?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-11 04:47:44
[cucumber/common] Comment on #1532
Quick check: berp in it's current form cannot be executed with .NET 5, so I need to recompile it to that platform. Unfortunately it has a dependency (for the template parser engine), that has not been updated to .NET 5. There are alternatives though, so it is possible to replace it with (hopefully) minimal changes in the razor files, but it needs a bit more time.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-11 07:17:34
[cucumber/common] Pull request merged by gasparnagy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-11 07:19:05
[cucumber/common] Pull request closed by renovate[bot]
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-11 09:07:06
[cucumber/common] Comment on #1517
Thanks <a href="https://github.com/Yellaks">@Yellaks</a>.
As you've seen we've already removed protobuf from the master branch. There is more work involved than making a new release of the messages library; All the Cucumber implementations must be refactored to use the new messages library.
As Cucumber is an open source project where most contributors contribute in their spare time we're unable to provide any time estimates.
If it's urgent to you, you could help by submitting a pull request to godog that uses the most recent (unreleased) version of messages, which are mirrored here: <a href="https://github.com/cucumber/messages-go">https://github.com/cucumber/messages-go</a>
Can we count on you to help with this?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-11 11:36:35
[cucumber/common] Comment on #1533
> <a href="https://github.com/thiagojacinto">@thiagojacinto</a> could you please clarify what you mean by that?
Sure, <a href="https://github.com/aslakhellesoy">@aslakhellesoy</a>. If you follow the same steps described by Frankie to reproduce this but adding the message formatter with `--format message:./report.json` the same thing happens:
the test fails on the first run, but passes on a retry, the `message` formatter incorrectly records it as a failed test, while the summary recording it as a success
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-11 13:48:23
[cucumber/common] Comment on #1534
OK, so as we'd expect, this is failing without mono
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-11 14:55:42
[cucumber/common] Pull request opened by aslakhellesoy
<p>This avoids running npm install when we specify e.g. <code>LANGUAGES=java make</code></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-11 21:51:11
[cucumber/common] Comment on #1535
Nice one, thanks!
Should we `rm -rf ../../node_modules` in the `clean` target too?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-12 01:31:06
[cucumber/common] Comment on #1535
We could, but I think I prefer to add a `clobber` target for that. It's nice to be able to `clean` just built files without deleting `node_modules`. Will push this change.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-12 01:33:31
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-12 02:34:54
[cucumber/common] Comment on #1534
This is passing now. We only run `mono` if the parser needs to be regenerated, but since it's committed to `git`, that's never necessary in CI. It is however needed for local development (in docker) whenever we change the Gherkin grammar.
So we need to manually check that we can regenerate the parser locally after a modification to `gherkin.razor`.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-12 03:12:11
[cucumber/common] Issue opened by gitissuepost
<p>Importing Classes from io.cucumber.messages.types gives error saying "the import cannot be resolved"</p>
<p>The same packages are used in some test classes of gherkin-java project. I tried importing that project and ended up with the same error.</p>
<p>import io.cucumber.messages.types.Envelope;<br />
import io.cucumber.messages.types.Feature;<br />
import io.cucumber.messages.types.GherkinDocument;<br />
import io.cucumber.messages.types.Pickle;<br />
import io.cucumber.messages.types.PickleStep;<br />
import io.cucumber.messages.types.Scenario;</p>
<p>• Version used: messages: 15.0.0, cucumber-core: 6.10.3
• Operating System and version: Windows 10</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-12 17:20:43
[cucumber/common] Comment on #1534
<a href="https://github.com/gasparnagy">@gasparnagy</a>'s take is that we only needed the `mono-runtime` package, not `mono-devel` which seems to have installed find on both architectures, so I think we have a good image now.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-12 21:02:38
[cucumber/common] Comment on #1534
Grrr, dunno why CircleCI won't start this container. I'll try pushing an image with a unique tag.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-12 21:49:13
[cucumber/common] Comment on #1534
I have no idea why CircleCI is failing the build now. It says
```
CircleCI was unable to run the job runner because we were unable to execute commands in build container.
This typically means that the build container entrypoint or command is terminating the container prematurely, or interfering with executed commands. Consider clearing entrypoint/command values and try again.
```
Nothing has changed about the container image that affects the entrypoint, and I can enter it via `make docker-run` and run commands in it fine on my local machine.
Anyone got ideas how to get past this?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 02:24:21
[cucumber/common] Comment on #1517
Yes, I can help with updating GoDog :)
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 05:12:10
[cucumber/common] Comment on #1532
Good news. I was able to upgrade berp... I want to make a few tests with cucumber, but it looks promising...
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 05:44:04
[cucumber/common] Issue reopened by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 05:44:04
[cucumber/common] Comment on #249
Reopening this because of discussion in <a href="https://cucumberbdd.slack.com/archives/C62D0FK0E/p1620900073467600">Slack</a>
<a href="https://github.com/davidjgoss">@davidjgoss</a>:
> What's the consensus across implementations about scenarios that contain no steps? This seems to break cucumber-js, not sure if it's worth addressing <a href="https://github.com/cucumber/cucumber-js/issues/1668">cucumber/cucumber-js#1668</a>
<a href="https://github.com/mpkorstanje">@mpkorstanje</a>:
> Well, Aslak seems to think it's a good idea to have the empty scenario result in an undefined result. I think that empty scenarios should be treated like an ordinary scenario. Execute all the before/after hooks, but not any steps (because there are none). It makes it much easier to keep everything consistent (esp when dealing with events, scenario outcomes, ect). If empty scenarios are undesirable that should be handled by a linter.
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a>:
> I don't have a strong opinion about it, just that we implement it consistently.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 06:28:27
[cucumber/common] Comment on #1534
Can you remove the formatting edits please?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 06:29:01
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 06:29:29
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 07:14:24
[cucumber/common] Comment on #1536
The types you are looking for are located in `import io.cucumber.messages.Messages.*`
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 07:14:24
[cucumber/common] Issue closed by mpkorstanje
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 08:17:21
[cucumber/common] Issue opened by mpkorstanje
<p>When Cucumber did not execute any scenarios the <code>io.cucumber:html_formatter:13.0.0</code> will display the message: <code>No match found for: "undefined"</code>.</p>
<p>To reproduce:</p>
<pre><code>package io.cucumber.examples.calculator;
import io.cucumber.junit.Cucumber;
import io.cucumber.junit.CucumberOptions;
import org.junit.runner.RunWith;
@RunWith(Cucumber.class)
@CucumberOptions(tags = "@no-such-tag", plugin = {
"html:target/results.html",
"message:target/results.ndjson"
})
public class RunCucumberTest {
}
</code></pre>
<p><code>target/results.html</code>:<br />
<a href="https://user-images.githubusercontent.com/6946919/118145685-23e9d180-b40e-11eb-8d8e-46fe1116f590.png">image</a></p>
<p><code>target/results.ndjson</code>:</p>
<pre><code>{"meta":{"protocolVersion":"15.0.0","implementation":{"name":"cucumber_jvm","version":"${parent.version}"},"runtime":{"name":"OpenJDK 64-Bit Server VM","version":"11.0.11+9_Ubuntu_0ubuntu2.20.04"},"os":{"name":"Linux"},"cpu":{"name":"amd64"}}}
{"testRunStarted":{"timestamp":{"seconds":"1620918593","nanos":632453000}}}
{"source":{"uri":"classpath:io/cucumber/examples/calculator/basic_arithmetic.feature","data":"@foo\nFeature: Basic Arithmetic\n\n Background: A Calculator\n Given a calculator I just turned on\n","mediaType":"text/x.cucumber.gherkin+plain"}}
{"gherkinDocument":{"uri":"classpath:io/cucumber/examples/calculator/basic_arithmetic.feature","feature":{"location":{"line":2,"column":1},"tags":[{"location":{"line":1,"column":1},"name":"@foo","id":"5f302c14-f235-4657-a9cf-a4d5a2d90771"}],"language":"en","keyword":"Feature","name":"Basic Arithmetic","children":[{"background":{"location":{"line":4,"column":3},"keyword":"Background","name":"A Calculator","steps":[{"location":{"line":5,"column":5},"keyword":"Given ","text":"a calculator I just turned on","id":"5abd1c65_c068_4479_ba0c_c3324a52bf4e"}],"id":"48e0b2c8-6339-485c-a4f2-51ef107de253"}}]}}}
{"testRunFinished":{"success":true,"timestamp":{"seconds":"1620918593","nanos":653527000}}}
</code></pre>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 08:31:25
[cucumber/common] Comment on #1534
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> done. I'm still none the wiser what's up with CircleCI here
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 08:40:10
[cucumber/common] Comment on #1534
Dunno - I can try to have a look tomorrow or next week
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 08:47:47
[cucumber/common] Comment on #1534
I'm scared to merge this in until we figure it out. I've asked CircleCI support about it.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 09:29:53
[cucumber/common] Comment on #1534
It’s obviousy not mergable :-)
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 10:20:33
[cucumber/common] Comment on #1534
I had accidentally pushed a tag that only had an `arm64` arch image. It's looking better now.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 11:39:25
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about these updates again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 11:40:30
[cucumber/common] Pull request merged by mpkorstanje
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 12:41:45
[cucumber/common] Comment on #1534
Great work <a href="https://github.com/mattwynne">@mattwynne</a>! What's left to do? Release 0.4.0 of the image and use that?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 12:46:20
[cucumber/common] Comment on #1539
The acceptance tests for `@cucumber/react` try to render all the message streams from the cck, so if we add this to the cck it should be easier to reproduce.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 12:47:52
[cucumber/common] Comment on #1539
This is vaguely related to <a href="https://github.com/cucumber/common/pull/1498">#1498</a>, but I think that's a different case.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 20:00:00
[cucumber/common] Comment on #1534
Yep! I'll merge <a href="https://github.com/cucumber/cucumber-build/pull/12">cucumber/cucumber-build#12</a> and make a 0.4.0 release. There are a couple more issues to work through but I think we're off the build platform and into the code now.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-13 21:42:49
[cucumber/common] Comment on #249
> I think that empty scenarios should be treated like an ordinary scenario. Execute all the before/after hooks, but not any steps (because there are none). It makes it much easier to keep everything consistent (esp when dealing with events, scenario outcomes, ect). If empty scenarios are undesirable that should be handled by a linter.
I like all of those words. My only question is whether or not steps from a `Background` count or not.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-14 05:22:04
[cucumber/common] Pull request ready for review by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-14 05:43:13
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-14 05:56:07
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-14 06:58:33
[cucumber/common] Pull request opened by gasparnagy
<p>Upgrading Berp to a version runs on .NET 5.0</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-14 08:12:06
[cucumber/common] Comment on #1507
Looks like someone went ahead and did this for Aruba too. Not against it per se, but a little surprised that I wasn't notified 😕
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-14 08:23:43
[cucumber/common] Comment on #1507
A pull request would have been nice since the Aruba documentation refers to `master` all over the place 😞.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-14 08:28:40
[cucumber/common] Pull request opened by aslakhellesoy
<p>It was previously hard-coded to true. The CCK is updated.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-14 10:27:12
[cucumber/common] Issue opened by gitissuepost
<p>The types you are looking for are located in <code>import io.cucumber.messages.Messages.**</code></p>
<p><em>Originally posted by <a href="https://github.com/mpkorstanje">@mpkorstanje</a> in <a href="https://github.com/cucumber/common/issues/1536#issuecomment-840590699">#1536 (comment)</a></em></p>
<p><code>Parser&lt;GherkinDocument&gt; parser = new Parser(new GherkinDocumentBuilder(idGenerator));</code></p>
<p>The above code ends up with exception while using these Classes from Messages class.</p>
<p>My Code:</p>
<p>`import io.cucumber.gherkin.GherkinDocumentBuilder;<br />
import io.cucumber.gherkin.Parser;<br />
import io.cucumber.gherkin.TokenMatcher;<br />
import io.cucumber.messages.IdGenerator;<br />
import io.cucumber.messages.Messages.GherkinDocument;</p>
<p>public class SampleGherkinDocumentBuilderTest {<br />
private final static IdGenerator idGenerator = new IdGenerator.Incrementing();</p>
<pre><code>public static void main(String[] args) {
Parser&lt;GherkinDocument&gt; parser = new Parser(new GherkinDocumentBuilder(idGenerator));
TokenMatcher matcher = new TokenMatcher();
GherkinDocument d1 = parser.parse("Feature: 1", matcher);
GherkinDocument d2 = parser.parse("Feature: 2", matcher);
System.out.println(d1.getFeature().getName());
System.out.println(d2.getFeature().getName());
}
</code></pre>
<p>}`</p>
<p>Help me understand what am I missing here as <code>Parser&lt;GherkinDocument&gt; parser = new Parser&lt;GherkinDocument&gt;(new GherkinDocumentBuilder(idGenerator));</code> ends up in compiler error</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-14 10:29:06
[cucumber/common] Issue closed by gitissuepost
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-14 10:32:29
[cucumber/common] Issue reopened by gitissuepost
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-14 14:35:02
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-14 14:35:52
[cucumber/common] Comment on #1542
Is this good to merge now? LGTM!
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-14 14:46:29
[cucumber/common] Comment on #1391
Hi <a href="https://github.com/davidjgoss">@davidjgoss</a> - I've brought this uptodate with a few big changes that have gone into master. What's left to do before we can merge this?
Maybe we can address tailwind and styling of the "chrome" in a separate PR?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 00:11:43
[cucumber/common] Comment on #1542
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> no, this is now only fixes it for dotnet. But a review is welcome. If you are fine with the approach, i will replicate it to the other languages.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 00:16:57
[cucumber/common] Comment on #1542
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> what do you think about creating a <a href="http://shared.mk">shared.mk</a> file that is a single file in the templates, but it is rsynced to all language folders. That way we could reduce the duplications in the language specific <a href="http://default.mk">default.mk</a> files...
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 05:56:04
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about these updates again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 09:02:58
[cucumber/common] Comment on #1391
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a>
Thanks for updating, and agreed on deferring work on the "chrome" to a future change.
I'm just working through the rest of the Gherkin component styling in terms of getting it looking good, encapsulating the styles and making it customisable - should be ready to go by Monday/Tuesday.
One question about `GherkinDocumentList` - do we think this needs to support custom rendering? It contains some business logic so I think if we wanted to allow users to replace it we'd probably want to break that stuff away from the presentational (accordion) stuff. Maybe we only allow custom CSS classes in this case?
BTW I deployed the Storybook output here so people can preview easily (I'll update it as I do more refinement):
<a href="https://serene-wilson-f8fc77.netlify.app/">https://serene-wilson-f8fc77.netlify.app/</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 10:07:43
[cucumber/common] Comment on #1542
We could, but it might come at the cost of additional complexity. Have a go at it if you want!
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 10:14:31
[cucumber/common] Comment on #1391
Personally I think I'd like to get rid of the accordion - I don't think it's an efficient way to navigate, especially with reports that have many files. You have to scroll too much to find what you want.
I think a scrollable tree navigation (Feature->Scenario->Step) on the left and a document on the right might work better. The index page (no document selected) could display a summary.
Not suggesting we do that in this PR, but if you agree we don't need to worry too much about GherkinDocumentList.
WDYT?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 10:20:29
[cucumber/common] Comment on #1544
Assuming you're using version 18.1.1 (the latest release as of today), you can find an example here:
<a href="https://github.com/cucumber/common/blob/gherkin/v18.1.1/gherkin/java/src/test/java/io/cucumber/gherkin/ParserTest.java">https://github.com/cucumber/common/blob/gherkin/v18.1.1/gherkin/java/src/test/java/io/cucumber/gherkin/ParserTest.java</a>
Please ask for this issue to be reopened if this doesn't solve your problem.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 10:20:29
[cucumber/common] Issue closed by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 10:37:04
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
cucumber/messages-go
<strong><a href="https://togithub.com/cucumber/messages-go/compare/v15.0.0...v16.0.0"><code>v16.0.0</code></a></strong></p>
<p><a href="https://togithub.com/cucumber/messages-go/compare/v15.0.0...v16.0.0">Compare Source</a></p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 10:45:05
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 10:45:05
[cucumber/common] Comment on #1513
Hi <a href="https://github.com/OrhanTozan">@OrhanTozan</a>,
Thanks for your making your first contribution to Cucumber, and welcome to the Cucumber committers team! You can now push directly to this repo and all other repos under the cucumber organization! 🍾
In return for this generous offer we hope you will:
• ✅ Continue to use branches and pull requests. When someone on the core team approves a pull request (yours or someone else's), you're welcome to merge it yourself.
• 💚 Commit to setting a good example by following and upholding our <a href="https://cucumber.io/conduct">code of conduct</a> in your interactions with other collaborators and users.
• 💬 Join the <a href="http://cucumber.io/support#slack">community Slack channel</a> to meet the rest of the team and make yourself at home.
• ℹ️ Don't feel obliged to help, just do what you can if you have the time and the energy.
• 🙋 Ask if you need anything. We're looking for feedback about how to make the project more welcoming, so please tell us!
On behalf of the Cucumber core team,
Aslak Hellesøy
Creator of Cucumber
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 10:47:18
[cucumber/common] Pull request closed by renovate[bot]
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 11:23:38
[cucumber/common] Pull request opened by aslakhellesoy
<p>After release of messages 16.0.0</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 12:02:30
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
benmosher/eslint-plugin-import
<strong><a href="https://togithub.com/benmosher/eslint-plugin-import/blob/master/CHANGELOG.md#%E2%80%8B2232---2021-05-15|<code">https://togithub.com/benmosher/eslint-plugin-import/blob/master/CHANGELOG.md#%E2%80%8B2232---2021-05-15|<code</a>v2.23.2</code>></strong></p>
<p><a href="https://togithub.com/benmosher/eslint-plugin-import/compare/v2.23.1...v2.23.2">Compare Source</a></p>
<p><strong>Changed</strong></p>
<p>• [meta] add <code>safe-publish-latest</code>; use <code>prepublishOnly</code> script for npm 7+</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled due to failing status checks.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 12:02:42
[cucumber/common] Comment on #1548
*Branch automerge failure*
This PR was configured for branch automerge, however this is not possible so it has been raised as a PR instead.
* * *
• Branch has one or more failed status checks
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 12:02:56
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 13:38:04
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 14:02:07
[cucumber/common] New release Release - cucumber-gherkin/v19.0.0 published by cukebot
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 14:08:04
[cucumber/common] Pull request closed by renovate[bot]
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 14:08:07
[cucumber/common] Pull request closed by renovate[bot]
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 15:17:22
[cucumber/common] Comment on #1317
<a href="https://github.com/a1dutch">@a1dutch</a> how would you like to specify additional metadata? How/where would you like it rendered?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 17:11:57
[cucumber/common] Pull request merged by renovate[bot]
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 21:40:58
[cucumber/common] Issue opened by mattwynne
<p>When we first moved to the monorepo, many of the package managers we used needed a library to be in it's own repo, so we set up individual read-only mirror repos which are copied to when there are changes in this repo.</p>
<p>The context has changed, and we don't need most of these any more. We might still need them for the go modules.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-15 22:24:48
[cucumber/common] Comment on #1544
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> : Can we please reopen this. I used the test to parse a feature file and it ended up with exception. Attached are a dump of my sample project and the exception I encounter. My requirement is to parse the feature file to manipulate that in run time.
E.g. I have an external source where all test data are being maintained (let's say Jira/ALM). Before the test gets triggered, we have capabilities to fetch those details from the source. After that we need to add test data fetched into example section and the resultant feature will be consumed by cucumber for execution. To update the Examples section, we need to build a parser and that's where I am looking for built-in classes, if any available, which can be utilized to parse the feature and re-create it with updates.
Initial Feature:
```
Feature: Sample Feature
Scenario: Sample Scenario
Given I launch chrome
When I navigate to <URL>
Then I will land to page with header <PageHeader>
```
Updated Feature to be re-created for execution:
```
Feature: Sample Feature
Scenario: Sample Scenario
Given I launch chrome
When I navigate to <URL>
Then I will land to page with header <PageHeader>
Examples:
|URL|PageHeader|
|<http://www.google.com>|Google|
```
<a href="https://github.com/cucumber/common/files/6488734/Sample-Parser.zip">Sample-Project Dump</a>
<a href="https://github.com/cucumber/common/files/6488735/Exception.Outcome.txt">Exception</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-16 02:07:55
[cucumber/common] Issue reopened by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-16 04:03:49
[cucumber/common] Comment on commit "Disable gherkin/elixir"
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> can I please get an explicit ping in the commit message? I'm monitoring the repository quite closely, but the number of failing builds is simply too overwhelming: there's lots of in-progress work which doesn't need my attention but does fail the build. I can't keep up this way.
With this commit message, I'm also unable to backtrack whether I should have taken action earlier as in that there's a branch waiting for my input. In that case, can I get an explicit ping on such a branch? Because .. (same as above).
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-16 05:38:13
[cucumber/common] Pull request opened by ehuelsmann
<p><strong>Summary</strong></p>
<p>Based on the example from Gherkin/dotNet, add Perl tests to the<br />
parallel build.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-16 05:43:49
[cucumber/common] Pull request opened by ehuelsmann
<p><strong>Summary</strong></p>
<p>The JSON schema specifies required output for some fields which used to be absent in the output under protobuf messages.</p>
<p><strong>Details</strong></p>
<p>Bring back compatibility and testability.</p>
<p><strong>Motivation and Context</strong></p>
<p>See JSON schema commits and discussion context.</p>
<p><strong>How Has This Been Tested?</strong></p>
<p>All regular tests are now passing again.</p>
<p><strong>Types of changes</strong></p>
<p>Compatibility with the spec.</p>
<p><strong>Checklist:</strong></p>
<p>☑︎ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-16 07:21:13
[cucumber/common] Pull request merged by ehuelsmann
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-16 08:17:48
[cucumber/common] Pull request merged by ehuelsmann
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-16 23:57:37
[cucumber/common] Comment on #1391
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> agreed re accordion, I've changed `GherkinDocumentList` to not be customisable.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 00:53:59
[cucumber/common] Comment on #249
We may let users choose how they would process their empty scenarios with an option:
• execute all hooks and background steps, and report the scenario with a status depending the execution of those background steps and hooks
• do not execute anything and report it as undefined skipped|empty|pending|...
I think both options may find their audience.
Also, I was curious to know how other tools consider empty tests. I have checked the behavior of rspec. It shows a "Not yet implemented" and report the test as pending. Hooks are not executed: if an error is raised, implemented tests are reported as "Error", but the not implemented test is still reported as pending.
```
$ bundle exec rspec spec/cucumber/cli/main_spec.rb:23 -fd
Run options: include {:locations=>{"./spec/cucumber/cli/main_spec.rb"=>[23]}}
Cucumber::Cli::Main
#execute!
passed an existing runtime
configures that runtime (FAILED - 1)
uses that runtime for running and reporting results (FAILED - 2)
do something (PENDING: Not yet implemented)
Pending: (Failures listed here are expected and do not affect your suite's status)
1) Cucumber::Cli::Main#execute! passed an existing runtime do something
# Not yet implemented
# ./spec/cucumber/cli/main_spec.rb:50
Failures:
1) Cucumber::Cli::Main#execute! passed an existing runtime configures that runtime
Failure/Error: raise "error"
RuntimeError:
error
# ./spec/cucumber/cli/main_spec.rb:10:in `block (2 levels) in <module:Cli>'
2) Cucumber::Cli::Main#execute! passed an existing runtime uses that runtime for running and reporting results
Failure/Error: raise "error"
RuntimeError:
error
# ./spec/cucumber/cli/main_spec.rb:10:in `block (2 levels) in <module:Cli>'
Finished in 0.00136 seconds (files took 0.75502 seconds to load)
3 examples, 2 failures, 1 pending
Failed examples:
rspec ./spec/cucumber/cli/main_spec.rb:30 # Cucumber::Cli::Main#execute! passed an existing runtime configures that runtime
rspec ./spec/cucumber/cli/main_spec.rb:40 # Cucumber::Cli::Main#execute! passed an existing runtime uses that runtime for running and reporting results
```
Do you know other tools which may be compared to cucumber as a BDD test runner?
I think we may comply to what is done with our "competitors" if some kind of consensus has already been found elsewhere.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 02:08:59
[cucumber/common] Comment on #249
While we could let users choose the behaviour, I don't think we should. It introduces complexity without significant benefit.
I think we should choose one behaviour and implement it consistently.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 03:13:12
[cucumber/common] Pull request opened by aslakhellesoy
<p>This fixes several bugs in create-meta where we weren't properly generating the right values.</p>
<p>This PR also introduces a new expression language - <code>${variable/pattern/replacement}</code> - similar to <a href="https://tldp.org/LDP/abs/html/parameter-substitution.html">bash parameter substitution</a>, but inspired from <a href="https://www.gnu.org/software/sed/manual/html_node/The-_0022s_0022-Command.html">sed's s command</a> which provides support for capture group back-references in the replacement.</p>
<p>This will make it easier to add support for other CI servers in the future, without any code changes.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 03:17:24
[cucumber/common] Comment on #249
For what it's worth, considering the empty scenario as a scenario with 0 steps that will execute before/after all and before/after scenario hooks and is passed by default is the most consistent and least complicated solution.
1. Hooks have access to the current scenario state. This can be elegantly defined as the worst result of all previous steps/hooks or passed if none.
2. So for consistency, after completion the scenario result can be defined as the worst result of all steps/hooks or passed if there was no result.
3. Undefined steps result in a suggested step definition snippet event. So for every undefined scenario there is at least one suggested snippet. By defining the scenario result as passed when there are no step (rather then undefined) we ensures that for an undefined step/scenario there is always at least one suggested snippet. This removes an edge case for implementers of plugins.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 03:19:21
[cucumber/common] Comment on commit "Disable gherkin/elixir"
Absolutely! Will do that the next time.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 03:49:15
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 04:23:00
[cucumber/common] New release Release - cucumber-gherkin/v19.0.1 published by cukebot
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 04:49:20
[cucumber/common] Comment on #1253
Fixed in <a href="https://github.com/cucumber/common/pull/1553">#1553</a> - released in 5.0.0
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 04:49:20
[cucumber/common] Issue closed by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 04:49:40
[cucumber/common] Issue closed by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 04:49:40
[cucumber/common] Comment on #1490
Fixed in <a href="https://github.com/cucumber/common/pull/1553">#1553</a> - released in 5.0.0
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 04:56:34
[cucumber/common] Comment on #249
The fact that a scenario is empty is an important information. It may even be an very important information in a BDD workflow, or within a living documentation. Reporting it as passed or failed is loosing that piece of information.
From a testing tool point of view, I know some tools which reports empty tests or tests without expectation, and requires the developer to explicitly specify that no assertion is actually expected. As annoying as it can be for the developer, it make it easier to retrieve that info. And for code maintenance, it also make it explicit that the test is empty or has no expectation on purpose.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 05:22:32
[cucumber/common] Comment on #1317
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> maybe move the search/filter down onto its own line and span the whole width, you could then add them in their own box on the top right (where the filter is).
Another option is to add them to the middle section (this might be pretty crowded, so it could be displayed in two columns and you could move the filter down as above).
you could pass these in as command line arguments
`-<a href="http://Dcucumber.plugin.react.info">Dcucumber.plugin.react.info</a>=App Version=1.2.3, App Commit=12da3fa2, Tested By=Andy`
or evironment variables
`CUCUMBER_PLUGIN_REACT_INFO=App Version=1.2.3, App Commit=12da3fa2, Tested By=Andy`
Cheers
Andy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 05:26:30
[cucumber/common] Comment on #1317
Ok, so if we use an environment variable like `CUCUMBER_PLUGIN_REACT_INFO=App Version=1.2.3, App Commit=12da3fa2, Tested By=Andy`, would you expect the value of that variable to be split on `,` and then rendered independently?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 05:36:31
[cucumber/common] Comment on #1317
yes seperated by `,` you might want to support quoted strings like below (but i could live without that), and then rendered indepently.
`CUCUMBER_PLUGIN_REACT_INFO=App Version="1f23,1d53", App Commit='1f23,1d53', Tested By="Holland, Andy"`
i dont know if there is a standard way to pass parameters to a plugin, cucumber-jvm only seems to accept one parameter as part of the plugin and this is the report location
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 05:42:42
[cucumber/common] Comment on #249
I'm thinking about the principle of least surprise for this one. As a user, given my intent is something like "putting this scenario title here as a placeholder, coming back later to flesh it out" (see early comments), I think I'd be more surprised if Cucumber ran the hooks and background steps.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 05:43:40
[cucumber/common] Comment on #1317
cucumber-js supports `--format-options` as a JSON literal but I think it might be the only implementation that does.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 05:52:16
[cucumber/common] Comment on #249
Following the principle of least surprise everything that happens before a scenario should happen before a scenario with no steps. The latter being a subset of the former.
As an analogue: Scenarios are the the equivalent of tests. Features the equivalent of a class with multiple tests. A test with no statements in the method body is equivalent to an empty scenario.
```
Feature: Test
Scenario: nothing
```
Would be equivalent to:
```
class Test {
@BeforeEach
public void before(){
}
@AfterEach
public void before(){
}
@Test
public void nothing(){
}
}
```
This is a valid junit test and will execute all hooks. If you have a decent linter, it will complain about a test without an assertion.
> a BDD workflow,
We can't and shouldn't try to facilitate all possible work flows during test execution. That's what linters are for. It also solves the problem of making things optional or not. Linters can usually be configured with rules about things they complain and don't complain about.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 05:57:32
[cucumber/common] Comment on #249
I don't really see a Scenario as being analogous to a JUnit test. My angle was that it only becomes a runnable test once I actually add some steps and implement the glue for them - before that its just a piece of unfinished documentation. This is maybe not the most prevalent point of view though...
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 06:06:01
[cucumber/common] New release Release - cucumber-json-formatter/v18.0.0 published by cukebot
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 06:17:18
[cucumber/common] Comment on #249
> My angle was that it only becomes a runnable test once I actually add some steps and implement the glue for them
So you are arguing that an empty scenario should not result in a pickle and completely ignored when it comes to execution and listing scenarios?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 06:30:39
[cucumber/common] Comment on #1521
Here's a first iteration on cucumber-ruby: <a href="https://github.com/cucumber/cucumber-ruby/pull/1542">cucumber/cucumber-ruby#1542</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 06:42:52
[cucumber/common] Comment on #249
For me, yes - but this does contradict <a href="https://github.com/mattwynne">@mattwynne</a> original point about not having visibility of these empty scenarios if they never manifest as pickles.
(Conscious that I'm not really adding anything new here, so going to ease up a bit :))
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 07:13:10
[cucumber/common] Comment on #249
> As an analogue: Scenarios are the the equivalent of tests. Features the equivalent of a class with multiple tests. A test with no statements in the method body is equivalent to an empty scenario.
>
> ```
> Feature: Test
> Scenario: nothing
>
> ```
>
> Would be equivalent to:
>
> ```
> class Test {
>
> @BeforeEach
> public void before(){
>
> }
>
> @AfterEach
> public void before(){
>
> }
>
> @Test
> public void nothing(){
>
> }
> }
>
> ```
>
> This is a valid junit test and will execute all hooks. If you have a decent linter, it will complain about a test without an assertion.
Yes, this is a valid and passing junit test.
But the equivalent is not a valid and passing `rspec`.
<a href="https://jsfiddle.net/tnLv25r7/">Mocha would also report it as a pending test</a>.
<a href="https://phpunit.readthedocs.io/en/9.5/risky-tests.html#useless-tests">PhpUnit also has a specific behavior for test without any assertion in it</a>.
So we should not rely only on how junit behaves to decide how cucumber should behave I guess.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 07:19:26
[cucumber/common] Comment on #249
I don't think your fiddle worked. It's not showing me a pending test.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 07:22:44
[cucumber/common] Comment on #249
> I don't think your fiddle worked. It's not showing me a pending test.
Fixed :D
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 07:43:48
[cucumber/common] Comment on #249
> PhpUnit also has a specific behavior for test without any assertion in it.
PHP Unit does this by checking after the test was executed if any assertions were invoked. This is essentially a build in after hook. PHP Unit can do this because they provide their own assertion library. In Java there are many to chose from (most linters just check for the word assert).
> Fixed :D
If you pass in an empty function, it does pass. 😆
```
mocha.setup("bdd");
chai.should();
describe("Test", function() {
it("nothing", function(){
});
});
mocha.run();
```
And mocha will execute hooks, even if the test is pending:
```
mocha.setup("bdd");
chai.should();
describe("Test", function() {
before(function() {
throw "oops"
});
it("nothing");
});
mocha.run();
```
Though marking the test as `pending` rather then `undefined` solves the problems with objection <a href="https://github.com/cucumber/common/issues/3">#3</a>. It does leave <a href="https://github.com/cucumber/common/issues/1">#1</a> and <a href="https://github.com/cucumber/common/issues/2">#2</a> in place.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 11:58:24
[cucumber/common] Pull request reopened by renovate[bot]
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 12:32:09
[cucumber/common] Comment on #1534
<a href="https://github.com/gasparnagy">@gasparnagy</a> <a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> any idea why the series build failed but the parallel one didn't? It looks like the problem in the series build was in building gherkin/go, and Berp couldn't find .NET:
```
if [[ -d go ]]; then cd go && make default; fi
make[2]: Entering directory '/home/cukebot/cucumber/gherkin/go'
mono /var/lib/berp/1.1.1/tools/net471/Berp.exe -g gherkin.berp -t parser.go.razor -o parser.go
Berp 1.1.1
Copyright (C) 2014-2020, Gaspar Nagy (Spec Solutions)
Licensed under the Apache License, Version 2.0 (the "License")
<http://www.apache.org/licenses/LICENSE-2.0>
Generating parser for grammar 'gherkin.berp' using template 'parser.go.razor'.
Loading grammar...
The grammar was loaded with 14 tokens and 19 rules.
Calculating parser states...
51 states calculated for the parser.
Generating parser class...
Generation failed.
Could not load file or assembly 'Microsoft.CSharp, Version=4.0.0.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a' or one of its dependencies.
make[2]: *** [Makefile:94: parser.go] Error 1
make[2]: Leaving directory '/home/cukebot/cucumber/gherkin/go'
make[1]: *** [<a href="http://default.mk:17">default.mk:17</a>: default-go] Error 2
make[1]: Leaving directory '/home/cukebot/cucumber/gherkin'
make: *** [Makefile:26: default-gherkin] Error 2
Exited with code exit status 2
```
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 13:11:02
[cucumber/common] Comment on #1534
I'm not sure why it failed.
What's more puzzling to me is why it ran in the first place. The generated `parser.go` is added to git, and unless its Makefile dependencies (`gherkin.berp` or `parser.go.razor`) have more recent timestamps, it shouldn't have to be regenerated.
Berp typically only needs to run on developer workstations when the grammar or template is modified.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 13:12:28
[cucumber/common] Comment on #1526
LGTM, but the `Makefile` also needs to be updated to use the same image. <a href="https://github.com/mattwynne">@mattwynne</a> ?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 14:38:40
[cucumber/common] Comment on #1526
Yeah this is basically a robotic dupe of <a href="https://github.com/cucumber/common/pull/1534">#1534</a> right?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 14:39:09
[cucumber/common] Pull request ready for review by mattwynne
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 14:41:26
[cucumber/common] Pull request closed by mattwynne
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 14:43:13
[cucumber/common] Comment on #1526
*Renovate Ignore Notification*
As this PR has been closed unmerged, Renovate will now ignore this update (0.4.4). You will still receive a PR once a newer version is released, so if you wish to permanently ignore this dependency, please add it to the `ignoreDeps` array of your renovate config.
If this PR was closed by mistake or you changed your mind, you can simply rename this PR and you will soon get a fresh replacement PR opened.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 15:03:56
[cucumber/common] Comment on #249
<a href="https://github.com/mpkorstanje">@mpkorstanje</a> there's something you said that I don't understand:
> Undefined steps result in a suggested step definition snippet event. So for every undefined scenario there is at least one suggested snippet.
I don't see how we'd have an undefined _step_ here since there are no steps. What did I miss?
Let's use a concrete example, we like those 😁
`login.feature`
```
Feature: Login
Background:
Given a user Dave
Scenario: Successful login
When Dave logs in with his password
Then he can access the dashboard
Scenario: Failed login
```
The `Failed login` scenario is blank because the team are still deciding how to handle a failed login attempt, so they leave it blank.
Let's imagine there's also a hook in this suite to reset the state of the database between tests.
`hooks.js`
```
const { Before } = require('@cucumber/cucumber')
Before(() => Database.reset)
```
What I'd like to have happen is that the team can see in their results that there are two scenarios, and one of them is "not finished". `undefined` seems ideal to me, but `pending` would also work:
```
2 scenarios (1 passed, 1 undefined)
```
I can see that considering the case of the background steps adds complexity - I don't know if we have information in the pickle steps about whether a step came from a background or not - so it might be nice to ignore that one for now. I do think we should deal with hooks though. I don't think this would be very useful otherwise.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 15:07:02
[cucumber/common] Comment on #249
I can imagine how we'd implement this in cucumber-ruby, once the pickle compiler is passing these empty scenarios through. Perhaps <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> and I should pair on this and we can show you what we come up with, for discussion.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 15:19:31
[cucumber/common] Comment on #1507
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> I'm wondering if it makes sense to do <a href="https://github.com/cucumber/common/issues/1550">#1550</a> before tackling the monorepo, as the subrepo stuff feels like the risky part of doing this on the monorepo, and the less of it there is to have to fix if it breaks, the better. WDYT?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 15:21:57
[cucumber/common] Comment on #1550
<a href="https://github.com/jaysonesmith">@jaysonesmith</a> or anyone else from <a href="https://github.com/orgs/cucumber/teams/go">@cucumber/go</a> can you help me understand whether we still need the subrepos for publishing common go modules?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 16:04:59
[cucumber/common] Pull request merged by mattwynne
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 16:36:56
[cucumber/common] Comment on #249
The proposal is the following rule:
• The scenario state is "undefined" if no steps have been executed.
This comes on top of the existing rule:
• The scenario state is the worst state of the executed steps.
I'm making a two pronged argument as to why this is a bad idea.
Prong 1:
For example you write a plugin that prints a summary of a scenario executions. To make this possible, whenever an undefined step is encountered a suggested snippet event is emitted. So there is a third rule:
• For undefined step there is an suggested snippet event.
So you could collect all these suggested snippets for each scenario, and when a scenario finishes with the status "undefined" you print the list of suggested snippets. BUT if there are no snippets, the scenario had no steps and a different message must be printed. And that's a really weird edge case to deal with when writing a plugin.
The reason you have this edge case is because there are two ways for a scenario to get the "undefined" state:
1. Either the execution of any of the steps end with the "undefined" state.
2. Or the scenario did not have any steps and defaulted to "undefined".
You could work around this by instead proposing:
• The scenario state is "pending" if there were no steps.
But this will run into problems the second prong.
Prong 2:
Before- and after-hooks have access to the current scenario state. This for example allows them to take screenshots when a scenario did not pass. Following the the proposed rule the first before hook will see the scenario state as "undefined" because no steps have yet been executed.
You could work around this by defining the default scenario state from the perspective of the consumer:
• The scenario state provided to a hook is "passed" if no steps have been executed.
• The scenario state provided to an event is "undefined" if no steps have been executed.
Implementation wise this is also kinda annoying because now depending on the consumer you have to calculate the state of the scenario differently. But more importantly this results in an inconsistency between hooks and event consumers when dealing with the empty scenario. The last after hook would see the scenario as passed while the event consumer would see the scenario as undefined.
And of course there are ways to work around this and add more and more nuance. However by doing this we keep adding more and more complexity. Complexity that we absolutely shouldn't have to deal with to finesse what amounts to an edge case. In nearly every other test framework in existence the empty test passes by definition.
Hence my proposal:
• The scenario state is "passed" if no steps have been executed.
• The scenario state is the worst state of the executed steps.
This allows us to treat the empty scenario as any other scenario, but with an empty list of steps. Before- and after-hooks are executed as the would for any other scenario with steps. Hooks and and event consumers can be build on the same assumptions. There is no weird interaction with suggested snippet events and the whole thing can be expressed elegantly as:
```
private final List<Result> stepResults = new ArrayList<>();
public Status getStatus() {
if (stepResults.isEmpty()) {
return Status.
}
return max(stepResults, comparing(Result::getStatus)).getStatus();
}
```
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 16:56:47
[cucumber/common] Comment on #249
> Let's use a concrete example, we like those.
> I can see that considering the case of the background steps adds complexity
So far I've been talking about compiled pickles with no steps. You've gone down a different rabit hole with your example:
For everything I wrote the following example can be used:
```
Feature: Login
Scenario: Successful login
Given a user Dave
When Dave logs in with his password
Then he can access the dashboard
Scenario: Failed login
```
Here the failed login scenario has an empty list of steps.
However for this example:
```
Feature: Login
Background:
Given a user Dave
Scenario: Successful login
When Dave logs in with his password
Then he can access the dashboard
Scenario: Failed login
```
The question should be: "Does the Failed login scenario compile into a pickle with 0 or 1 steps".
I'm somewhat ambivalent about either. With what I've been proposing so far regardless if the scenario has 1 or 0 steps, the outcome would probably be passed. However with the other proposal the outcome might also be undefined/pending. And then this distinction matters because it would flag a potential problem in the workflow.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 17:10:35
[cucumber/common] Comment on #249
Anyway, I get the impression that all arguments for pending/undefined stem from the need to facilitate a test driven workflows where scenarios are intentionally incomplete.
However we are discussing Cucumbers execution model and implicitly its event model. We are also trying to keep it consistent across multiple implementations and integrate it with various test runners and IDEs. What ever complexity we add here is expensive, time consuming and very hard to change. This means we need simple and well defined behaviors.
Yet workflows are fuzzy and mostly involve people. This means we need configuration, flexibility and nuance. The execution model is the wrong abstraction layer to address these concerns. By the time we get this deep into Cucumbers guts workflow issues should have been dealt with.
This is why I've been proposing a linter. Linters are the ideal tool to handle workflows. They don't require Cucumber to be executed. This makes them light weight in terms of dependencies and allows them to run while writing features. This removes a lot of complexity from the linter and potentially makes them language agnostic.
Linters tend to come with configurable rules and a set of of defaults. This makes them ideal for dealing with individual workflows. Linters usually also tend to have machine consumable output for easy integration with IDE's and other tools. All things we can't really cram into Cucumbers execution model and shouldn't try.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 17:42:01
[cucumber/common] Comment on #249
> What did I miss?
I'd say about 2-3 years of working on Cucumber. 😛
Like Cucumber Ruby, Cucumber JVM used to default to "Undefined" for empty scenarios. Along with strict/lenient-mode it has been an endless source of bugs and confusion. So I removed both from Cucumber JVM. I did this by defaulting to strict mode, removing lenient mode and treating empty scenarios as passed by definition. Now empty scenarios never fail the build while undefined scenarios always fail the build.
In practice it hasn't made a lick of a difference for the end user. In the JVM ecosystem tests frameworks only support three test outcomes; passed, skipped, or failed. And in CI this is reduced to either passed or failed. So Cucumbers very nuanced test outcome with six possible states is ultimately reduced to a binary or trinary outcome.
Simply not bothering with some of the nuance removed a bucket of complexity. This has yet to be a source of complaints. The bugs and inconsistencies on the other hand were. So I think it is important to look at this holistically.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-17 23:54:16
[cucumber/common] Comment on #249
> In practice it hasn't made a lick of a difference for the end user. In the JVM ecosystem tests frameworks only support three test outcomes; passed, skipped, or failed. And in CI this is reduced to either passed or failed. So Cucumbers very nuanced test outcome with six possible states is ultimately reduced to a binary or trinary outcome.
>
> Simply not bothering with some of the nuance removed a bucket of complexity. This has yet to be a source of complaints. The bugs and inconsistencies on the other hand were. So I think it is important to look at this holistically.
Good point here: maybe we should not try harder to report empty scenarios consistently. As you said, it may depend on the ecosystem.
As it would make perfect sense to have empty scenarios reported as "pending" in ruby and JS, if it make absolutely no sense in Java (even the "skipped" one?), maybe we should stop trying do have the exact same behavior for it.
That should not prevent the common packages to be consistent as the definitive status of a test remain hardly tied to the implementation itself
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-18 00:07:02
[cucumber/common] Comment on #1542
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> <a href="https://github.com/mattwynne">@mattwynne</a> how can it be that gherkin-java does not understand "dotnet" is that running on a different docker container?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-18 02:37:08
[cucumber/common] Comment on #1542
<a href="https://github.com/gasparnagy">@gasparnagy</a> yes:
<a href="https://github.com/cucumber/common/blob/fff7cc2c1a740563a8615b5e9042c1daa881fb64/.circleci/config.yml#L609-L610">common/.circleci/config.yml</a>
Lines 609 to 610 in </cucumber/common/commit/fff7cc2c1a740563a8615b5e9042c1daa881fb64|fff7cc2>
Also see my comment here: <a href="https://github.com/cucumber/common/pull/1534#issuecomment-842598720">#1534 (comment)</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-18 02:59:40
[cucumber/common] Comment on #1542
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> I see. So probably this was working so far, because the parser was never outdated and therefore berp wasn't invoked. But since I introduced the `.berp_restored` target, that runs even in case the parser is up-to-date.
Like here: <a href="https://github.com/cucumber/common/pull/1542/files#diff-fc82761c3f102c231bc696c79cacd60ef5e64fc6452ed707682cfc4b27206ceaR18">https://github.com/cucumber/common/pull/1542/files#diff-fc82761c3f102c231bc696c79cacd60ef5e64fc6452ed707682cfc4b27206ceaR18</a>
Is there an easy way in make to say that it should only invoke the target if the other targets are outdated?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-18 04:59:30
[cucumber/common] Comment on #1542
> Is there an easy way in make to say that it should only invoke the target if the other targets are outdated?
Yes, by using Make the way it was meant to be used :-)
Make targets should be files, not abstract names.
Here is an example:
```
foo:
echo "hello"
```
Running `make foo` will execute the target body _if and only if_ the `foo` file does not exist. Since the target never actually generates the `foo` file, it will _always_ run. This is not how make should be used. Let's fix it:
```
foo:
echo "hello" > $@
```
(The `$@` variable is the name of the target). When you run `make foo`, it will run the next time, but not the second, because `foo` now exists.
Let's get more advanced. The `foo` target now depends on the `bar` target:
```
foo: bar
echo "hello" > $@
car bar > $@
```
Create the `bar` file manually with `touch bar`. Then run `make foo`. It will run the 1st time but not the 2nd time, as expected.
Now, if you `touch bar` again, the timestamp of `bar` will be more recent than that of `foo`, so running `make foo` again will run the target again.
I hope that helps. I can have a look at this tomorrow if you haven't figured it out.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-18 05:22:25
[cucumber/common] Pull request opened by aslakhellesoy
<p>This fixes a bug in fake-cucumber where undefined steps were reported as SKIPPED when the previous step was not PASSED.</p>
<p>In the process I also fixed some eslint issues across the repo and made CI<br />
run eslint.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-18 06:10:42
[cucumber/common] Comment on #1542
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> Thx. At that level I was OK, but here I need to make sure that berp is installed... that is not a classic target, because it is only needed for another target to be established. The "make" way would be (base on my understanding):
```
parser.java: gherkin.berp template.razor /usr/bin/berp
berp ...
/usr/bin/berp:
...install berp
```
This is more or less what I have now, but this will attempt to install berp even if the parser is up-to-date (on eg. the java parallel build, where it is not possible). I tried now with some tricky `$eval/$call` magic, but I get a stupid error message `*** recipe commences before first target. Stop.`, so I kind of given up. If you have time tomorrow for a a quick pairing, I would be happy to fix it together (so that I can better learn from it).
Or I just give up using .NET restore for berp and upload it as a linux binary somewhere... But I thought it is easier to maintain the versions this way...
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-18 06:10:52
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-18 08:26:45
[cucumber/common] Comment on #1542
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> I think I have found a solution using a canned recipe for berp. <a href="https://github.com/cucumber/common/pull/1542/files#diff-b20a2c0f8de33a0b6b48411e3a44c33a583536f315050f81229be990eb42657eR70-R73">https://github.com/cucumber/common/pull/1542/files#diff-b20a2c0f8de33a0b6b48411e3a44c33a583536f315050f81229be990eb42657eR70-R73</a>
And the usage is simple now:
<a href="https://github.com/cucumber/common/pull/1542/files#diff-fc82761c3f102c231bc696c79cacd60ef5e64fc6452ed707682cfc4b27206ceaR18-R19">https://github.com/cucumber/common/pull/1542/files#diff-fc82761c3f102c231bc696c79cacd60ef5e64fc6452ed707682cfc4b27206ceaR18-R19</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-18 10:05:34
[cucumber/common] Comment on #1542
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> for some of the languages, the parser generation fails with the new berp version and so I need to fix those still:
• ruby
• python
• perl
• objective-c
(It only passes now because the parsers are not outdated.)
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-18 13:45:10
[cucumber/common] Issue opened by pgundlupetvenkatesh
<p>json-formatter: Threat Detected on cucumber-json-formatter-windows-amd64 by VirusTotal third-party online tool</p>
<p><strong>Summary</strong></p>
<p>My employer recently notified me not to use cucumber-json-formatter-windows-amd64 executable as it's unsafe by VirusTotal as it scanned and determined it as a potential threat.</p>
<p><strong>Detected Threats</strong></p>
<p>Please click the below link to view the threat details</p>
<p><a href="https://www.virustotal.com/gui/file/0fc4a7f15a78b65648e762b77069dc420c6dc331f172f2f7fe63c1ebeeaad71a/detection">https://www.virustotal.com/gui/file/0fc4a7f15a78b65648e762b77069dc420c6dc331f172f2f7fe63c1ebeeaad71a/detection</a></p>
<p><strong>Context & Motivation</strong></p>
<p>I am trying to fix the below warning using Cucumber Standalone JSON Formatter which will be consumed by the Cucumber Reports Jenkins plugin.</p>
<pre><code>WARNING: --format=json is deprecated and will be removed after version 5.0.0.
Please use --format=message and stand-alone json-formatter.
json-formatter homepage: <https://github.com/cucumber/cucumber/tree/master/json-formatter#cucumber-json-formatter>.
</code></pre>
<p><strong>Your Environment</strong></p>
<p>• Version used: Cucumber 4.1.0, ruby 2.6.3p62
• Operating System and version: Windows 10 Enterprise 2016</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 00:49:48
[cucumber/common] Comment on #1555
Maybe we could rename that `s` variable in the process?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 00:49:48
[cucumber/common] Comment on #1555
Maybe we could add a default value 'gherkin' to `syntax` to avoid breaking changes to the interface?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 00:49:49
[cucumber/common] aurelien-reeves commented on aslakhellesoy's pull request
I would have created two distinct renderers and call the good one from the pretty method depending the desired syntax.
It may look like duplication as the resulting code would look really close, but the intent behind it is actually distinct.
To illustrate that, I think of another format totally different from those two: another pretifier for it would look totally different, so having distinctive renderers would make more sense.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 01:01:26
[cucumber/common] Comment on #1555
> I would have created two distinct renderers and call the good one from the pretty method depending the desired syntax.
I did consider that, but went this solution to avoid too much code duplication. The traversal code is identical, and the only difference is prefixing/indentation of keywords. I think that can be improved by having distinct helper functions for each rendering function rather than using `if` statements.
I'll have a go at that.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 01:01:34
[cucumber/common] Comment on #1555
Agreed
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 01:03:05
[cucumber/common] Comment on #1555
It's an internal library, so I'm not worried about breaking the interface. I always avoid default argument values unless it's for keeping backwards compatibility (in public interfaces).
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 01:09:41
[cucumber/common] Comment on #1555
The library is published on npm publicly, and the pretty method is publicly exposed
Even if it is not documented, some developer may use it so it remains a breaking change of the public API IMO
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 01:49:39
[cucumber/common] Pull request ready for review by gasparnagy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 02:49:26
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
webpack/webpack
<strong><a href="https://togithub.com/webpack/webpack/releases/v5.37.1"><code>v5.37.1</code></a></strong></p>
<p><a href="https://togithub.com/webpack/webpack/compare/v5.37.0...v5.37.1">Compare Source</a></p>
<p><strong>Bugfixes</strong></p>
<p>• When using multiple configurations in watch mode and calling <code>Watching.invalidate</code>, <code>dependencies</code> and <code>parallelism</code> of the config array is now respected correctly
• Fix a crash when accessing the <code>stats</code> after the next compilation has started
• fix collecting changes when using <code>Watching.suspend</code>
• fix schema of <code>RuleCondition.not</code> and allow passing a condition directly instead of only an array</p>
<p><strong>Developer Experience</strong></p>
<p>• typings accept a ReadonlyArray of configurations now</p>
<p><strong>Contributing</strong></p>
<p>• fix coverage reporting for child processes
• remove outdated loader from readme</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled due to failing status checks.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 02:49:38
[cucumber/common] Comment on #1557
*Branch automerge failure*
This PR was configured for branch automerge, however this is not possible so it has been raised as a PR instead.
* * *
• Branch has one or more failed status checks
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 03:37:04
[cucumber/common] Pull request merged by gasparnagy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 04:08:31
[cucumber/common] Comment on #1542
🥳
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 04:36:16
[cucumber/common] Pull request merged by renovate[bot]
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 04:42:16
[cucumber/common] Comment on #1556
We've decided not to deprecate the JSON formatter since it's so widely used. If you upgrade to 6.1.0 that deprecation warning will be gone. See <a href="https://github.com/cucumber/cucumber-ruby/blob/main/CHANGELOG.md#changed-1">https://github.com/cucumber/cucumber-ruby/blob/main/CHANGELOG.md#changed-1</a> for details.
Regarding the virus scanner - it's a false positive.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 04:42:17
[cucumber/common] Issue closed by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 04:42:54
[cucumber/common] Comment on #1555
Please remember to label issues and pull requests :-)
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 04:43:24
[cucumber/common] Comment on #1532
Is this now fixed <a href="https://github.com/gasparnagy">@gasparnagy</a> ?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 04:45:47
[cucumber/common] Comment on #1532
Partly. <a href="https://github.com/cucumber/common/pull/1542">#1542</a> removed the reference to mono from the monorepo, so now we can also remove it from the build docker image, that is yet to be done.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 04:46:29
[cucumber/common] New release Release - cucumber-gherkin/v19.0.2 published by cukebot
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 05:02:05
[cucumber/common] Pull request opened by gasparnagy
<p>Added berp upgrade to gherkin v19.0.2 changelog</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 05:05:13
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 05:05:30
[cucumber/common] Comment on #1558
For small changes like this, please commit straight to the master branch.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 05:13:35
[cucumber/common] Comment on #1558
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> tried, but the GitHub web UI didn't allow me to do that (protected branch). I didn't try it directly from git though...
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 06:20:06
[cucumber/common] Comment on #1391
Almost done here. Left to do:
☐ Cleanup of a few last styling and encapsulation issues
☐ Add some tests to cover slight changes in logic in some components
☐ Add the ability for custom components to bail out and fall back to the default rendering - e.g. if I want to provide a custom implementation for a particular kind of attachment, but carry on with defaults for images etc
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 11:07:17
[cucumber/common] Comment on #1555
> I did consider that, but went this solution to avoid too much code duplication. The traversal code is identical, and the only difference is prefixing/indentation of keywords. I think that can be improved by having distinct helper functions for each rendering function rather than using `if` statements.
This reminds me of all the visitor patterns I did already for Gherkin AST... I wanted to make a generic one into Gherkin, but never got the time for that. I wonder how much of the duplication we could avoid with a generic gherkin AST visitor...
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 12:52:07
[cucumber/common] Comment on #249
Speaking of concrete examples, this is a reminder that, whatever is eventually decided, the test data will need to be updated (and, unless I am misinterpreting that expected output, the test has been inaccurate for years).
<a href="https://github.com/cucumber/common/blob/master/gherkin/testdata/good/incomplete_scenario.feature">https://github.com/cucumber/common/blob/master/gherkin/testdata/good/incomplete_scenario.feature</a>
<a href="https://github.com/cucumber/common/blob/master/gherkin/testdata/good/incomplete_scenario.feature.pickles.ndjson">https://github.com/cucumber/common/blob/master/gherkin/testdata/good/incomplete_scenario.feature.pickles.ndjson</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 14:25:20
[cucumber/common] Comment on #1391
`This comment was deleted`
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 14:54:42
[cucumber/common] Comment on #1391
I had to add the `cucumber-steps` class here in order to get the `list-style: none` CSS rule, but the drawback is that _any_ `<li>` now gets that style, including list items that are _not_ Gherkin steps.
I wonder if it would be better to move this style up to the dom element that renders the enclosing `Scenario`. That way the style would only apply to Gherkin Step list items.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 14:54:42
[cucumber/common] Comment on #1391
I think we should rename these styles to `./Cucumber.module.css`
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 14:54:43
[cucumber/common] Comment on #1391
This will render the step _as well as its DataTable_. This results in the Markdown table being rendered twice - once by the `<Step>` component, and once by the Markdown renderer (where it doesn't get any styles).
<a href="https://user-images.githubusercontent.com/1000/118888781-01672500-b8f4-11eb-8d64-9e661013f1d4.png">Screenshot 2021-05-19 at 22 39 07</a>
In order to avoid this double rendering I think it would be better to lift the `<DataTable>` rendering out of the `<Step>` component and have `<StepList>` render `<DataTable>` underneath the `<Step>`.
In Markdown mode, we'd let Markdown handle the rendering, but we could register custom `td`, `th` and `tr` compoents in `react-markdown`. We should consider extracting `Td`, `Th` and `Tr` components from `DataTable` so they can use the same components (and get the same styles).
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 14:54:44
[cucumber/common] aslakhellesoy requested changes on davidjgoss's pull request
I took the liberty to add a new `<MDG>` component that renders <a href="https://github.com/cucumber/common/blob/master/gherkin/MARKDOWN_WITH_GHERKIN.md">Markdown with Gherkin</a>.
That's highlighted some issues that I think it would be good to deal with. We don't necessarily need to deal with them in this PR - it could be done in a different PR.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 17:06:05
[cucumber/common] Comment on #1391
Addressed by changing to
```
.cucumber-steps {
padding: 0;
margin: 1em;
> li {
list-style: none;
}
}
```
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 17:06:41
[cucumber/common] Comment on #1391
Do we support using ordered lists as well with MDG?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 17:13:20
[cucumber/common] Comment on #1391
I see what you mean. I've pulled those baseline styles back out to the global stylesheet for now (and left the ones that apply to the accordion where they are). I think it's all stuff we might cover with Tailwind once we bring that in, so not going to fuss too much over the encapsulation for now.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 17:18:33
[cucumber/common] Comment on #1391
Oh interesting. I guess we'd have a similar situation for a doc string - a duplicate `<pre>`?
Pulling them out of the `<Step>` would help the Markdown problem, but would also change the order of things so that the data table or doc string would appear _after_ any attachments or errors for the step - I think this might be a bit confusing to read.
Coming at it from the opposite angle, what if we were to, for `table` and `pre`, check if the next non-whitespace line above is a list item (or maybe even just if we're within a list?), and if so just render nothing?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 17:27:11
[cucumber/common] Comment on #1391
FWIW pushed a commit that applies our data table style to react-markdown's table rendering.
Although looking at it now, it being presented like a Gherkin table but in a Markdown document feels a bit off 😬
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 17:37:46
[cucumber/common] Pull request ready for review by davidjgoss
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 22:15:14
[cucumber/common] Issue opened by enkessler
<p><strong>Summary</strong></p>
<p>I'm in the process of updating one of my projects to be able to use the 19.x major version of the <code>cucumber-gherkin</code> gem and am running into some unexpected API changes that do not appear to be reflected in the <a href="https://github.com/cucumber/common/blob/master/gherkin/CHANGELOG.md">CHANGELOG</a> at all.</p>
<p><strong>Expected Behavior</strong></p>
<p>Because the CHANGELOG only mentions some dialect updates (which would not impact my project) and dependency upgrades (whose own changelogs, in turn, had no impacting call outs), I was expecting things to still 'just work' after I upgraded from 18.x to 19.x.</p>
<p><strong>Current Behavior</strong></p>
<p>I'm seeing several places in the code where data keys in returned AST data got converted from <code>snake_case</code> to <code>camelCase</code> which, naturally, breaks any code that uses that particular portion of the API. So far, I've run into</p>
<p><code>parse_error</code> -> <code>parseError</code><br />
<code>media_type</code> -> <code>mediaType</code><br />
<code>doc_string</code> -> <code>docString</code><br />
<code>data_table</code> -> <code>dataTable</code></p>
<p>but I'm still in the process of upgrading so there may be several more that I'll run into.</p>
<p><strong>Possible Solution</strong></p>
<p>It depends on if all of the name changes were intentional or not. It's not the first time that the names of things have changed but I don't know if it was an attempt at standardization across language implementations or it's just that whoever wrote the code was in the habit of writing in a different style.</p>
<p>If the changes were intentional</p>
<p>• Update the CHANGELOG to reflect the changes</p>
<p>If the changes were not intentional</p>
<p>• Revert them and probably add some more testing around what the AST structure is so that these kinds of bugs don't slip through in the future</p>
<p><strong>Context & Motivation</strong></p>
<p>While I understand that an increase in the major version number indicates backwards incompatibility in some fashion, those incompatibilities tend to get called out. Since they weren't mentioned, I figured that I'd check and see if the changes were just as much of a surprise to everyone else as they were to me.</p>
<p><strong>Your Environment</strong></p>
<p>• Version used: <code>cucumber-gherkin (19.0.2)</code>
• Operating System and version: N/A
• Link to your project: <a href="https://github.com/enkessler/cuke_modeler">https://github.com/enkessler/cuke_modeler</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-19 23:57:16
[cucumber/common] Comment on #1550
We can easily specify a go module which is in a monorepo:
<a href="https://golang.org/ref/mod#module-path">https://golang.org/ref/mod#module-path</a>
<a href="https://golang.org/ref/mod#vcs-dir">https://golang.org/ref/mod#vcs-dir</a>
So in our case, that should be as simple as something like the following: <a href="http://github.com/cucumber/common/messages/go/v42">github.com/cucumber/common/messages/go/v42</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 00:13:37
[cucumber/common] Comment on #1559
Hello <a href="https://github.com/enkessler">@enkessler</a>
Thanks for the report.
This is actually related to messages@16.0. It does not use protocol buffers anymore, but jsonschema to describe the messages.
The impact you notice on AST is more a side effect than an expected change.
We need to improve the documentation, and the changelog, accordingly.
Regarding Ruby, the big difference is that messages are no longer classes, but hashes, and the key are now camelCase as you described.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 01:06:09
[cucumber/common] Pull request opened by aurelien-reeves
<p>Refs. <a href="https://github.com/cucumber/common/issues/1559">#1559</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 01:34:20
[cucumber/common] Comment on #1391
You mean this?
```
1. one
1. two
1. three
```
Which renders as:
1. one
2. two
3. three
MDG doesn't currently support it, but it definitely should! I'll add support for that
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 01:43:18
[cucumber/common] Comment on #1391
> ... but would also change the order of things so that the data table or doc string would appear after any attachments or errors for the step
Couldn't we address that by also lifting `<Attachment>` out of `<Step>`?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 01:56:20
[cucumber/common] Comment on #1391
Okay, so `<Step>` just becomes about the title line, the the other bits become other items in the list. Confused about how we'd still be able to maintain the order for Markdown though?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 03:34:15
[cucumber/common] Comment on #1391
Do you mean how we'd be able to display the error messages and attachments (runtime content) in Markdown?
I wonder if we could add some state to the `table` and `li` components and render runtime content directly after the `<Step>` (if there is no DataTable/DocString), otherwise after the DataTable/DocString.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 04:53:25
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 04:53:41
[cucumber/common] Comment on #1560
Thanks <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> - for small changes like this, please commit straight to master.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 04:58:53
[cucumber/common] Comment on #1517
Hi <a href="https://github.com/Yellaks">@Yellaks</a> - messages v16 has been released. Some of the structs have new names, have a look at <a href="https://github.com/cucumber/common/compare/gherkin/v18.1.1...gherkin/v19.0.0#diff-8cbcec67b5ee5b8964d2caab991e1d654096130cdd4e3ea0d46c7446b03df014">this huge diff</a> to get a sense of the kind of changes.
Thanks for offering to update godog!
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 04:58:54
[cucumber/common] Issue closed by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 05:30:52
[cucumber/common] Comment on #1560
I wanted to be sure not to have written mistakes in those changelogs
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 05:32:38
[cucumber/common] Comment on #1559
CHANGELOGs have been updated
Thanks again for your report
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 05:32:39
[cucumber/common] Issue closed by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 07:10:30
[cucumber/common] Comment on #1556
Also this: <a href="https://golang.org/doc/faq#virus">https://golang.org/doc/faq#virus</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 09:13:12
[cucumber/common] Comment on #1559
> This is actually related to messages@16.0
Yeah, I thought that might be the case but I didn't see anything in its CHANGELOG when I looked and when I searched in this repository for the new names, I found them under the `gherkin` sub-project, so I figured that this is where they came from (even if those changes were in response to something else further down the stack). In any case, I've now got my answer: It was intentional.
Thanks for the info and for updating the relevant CHANGELOG!
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 09:27:12
[cucumber/common] Comment on #1556
> Also this: <a href="https://golang.org/doc/faq#virus">https://golang.org/doc/faq#virus</a>
That's helpful, thanks.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 11:53:22
[cucumber/common] Comment on #1386
There are some docs that need to be updated, around <a href="https://github.com/cucumber/common/tree/master/messages#protocol-buffers">here</a>.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 11:58:37
[cucumber/common] Comment on #1521
Slightly tangential, but I learned about this feature of GitHub today: <a href="https://docs.github.com/en/communities/setting-up-your-project-for-healthy-contributions/creating-a-default-community-health-file">https://docs.github.com/en/communities/setting-up-your-project-for-healthy-contributions/creating-a-default-community-health-file</a>
e.g. <a href="https://github.com/Homebrew/.github">https://github.com/Homebrew/.github</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 12:06:11
[cucumber/common] Pull request opened by mattwynne
<p><strong>Summary</strong></p>
<p>Rename references to the master branch so we can rename it to main</p>
<p><strong>Motivation and Context</strong></p>
<p>The use of the word "master" as the default branch suggests we don't care about racism, which we do.</p>
<p><strong>How Has This Been Tested?</strong></p>
<p>It hasn't, and isn't easy to test. I'll fix it if something breaks.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 12:11:58
[cucumber/common] Comment on #1561
Merged in <a href="https://github.com/cucumber/common/commit/b94ce625967581de78d0fc32d84c35b46aa5a075">b94ce62</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 12:12:03
[cucumber/common] Pull request closed by mattwynne
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 12:29:21
[cucumber/common] Comment on #1507
These are the remaining repos that have `master` as default branch:
<a href="https://github.com/cucumber/aruba-getting-started">https://github.com/cucumber/aruba-getting-started</a>
<a href="https://github.com/cucumber/aruba-website">https://github.com/cucumber/aruba-website</a>
<a href="https://github.com/cucumber/commitbit">https://github.com/cucumber/commitbit</a>
<a href="https://github.com/cucumber/cucumber-android">https://github.com/cucumber/cucumber-android</a>
<a href="https://github.com/cucumber/cucumber-classroom">https://github.com/cucumber/cucumber-classroom</a>
<a href="https://github.com/cucumber/cucumber-cpp">https://github.com/cucumber/cucumber-cpp</a>
<a href="https://github.com/cucumber/cucumber-eclipse">https://github.com/cucumber/cucumber-eclipse</a>
<a href="https://github.com/cucumber/cucumber-eclipse-update-site">https://github.com/cucumber/cucumber-eclipse-update-site</a>
<a href="https://github.com/cucumber/cucumber-eclipse-update-site-snapshot">https://github.com/cucumber/cucumber-eclipse-update-site-snapshot</a>
<a href="https://github.com/cucumber/cucumber-electron">https://github.com/cucumber/cucumber-electron</a>
<a href="https://github.com/cucumber/cucumber-js-package-upgrade">https://github.com/cucumber/cucumber-js-package-upgrade</a>
<a href="https://github.com/cucumber/cucumber-json-testdata-generator">https://github.com/cucumber/cucumber-json-testdata-generator</a>
<a href="https://github.com/cucumber/cucumber-lua">https://github.com/cucumber/cucumber-lua</a>
<a href="https://github.com/cucumber/cucumber-message-upload-service">https://github.com/cucumber/cucumber-message-upload-service</a>
<a href="https://github.com/cucumber/cucumber-ruby-tcl">https://github.com/cucumber/cucumber-ruby-tcl</a>
<a href="https://github.com/cucumber/cucumber-tcl-wire">https://github.com/cucumber/cucumber-tcl-wire</a>
<a href="https://github.com/cucumber/cucumber.io-file-generator">https://github.com/cucumber/cucumber.io-file-generator</a>
<a href="https://github.com/cucumber/cucumber.ml">https://github.com/cucumber/cucumber.ml</a>
<a href="https://github.com/cucumber/git-subrepo">https://github.com/cucumber/git-subrepo</a>
<a href="https://github.com/cucumber/github-issue-stats">https://github.com/cucumber/github-issue-stats</a>
<a href="https://github.com/cucumber/microdata">https://github.com/cucumber/microdata</a>
<a href="https://github.com/cucumber/multi_test">https://github.com/cucumber/multi_test</a>
<a href="https://github.com/cucumber/protobuf">https://github.com/cucumber/protobuf</a>
<a href="https://github.com/cucumber/protobuf.js">https://github.com/cucumber/protobuf.js</a>
<a href="https://github.com/cucumber/repo-template">https://github.com/cucumber/repo-template</a>
<a href="https://github.com/cucumber/slack-invite-automation">https://github.com/cucumber/slack-invite-automation</a>
<a href="https://github.com/cucumber/todo-react-typescript-subsecond">https://github.com/cucumber/todo-react-typescript-subsecond</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-20 16:00:50
[cucumber/common] Comment on commit "Rename master to main #1507"
Thanks for doing this <a href="https://github.com/mattwynne">@mattwynne</a>!
Can you please fix your editor? Inadvertent formatting edits in your commits is a recurring thing.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-21 14:17:31
[cucumber/common] Comment on #1507
cucumber-electron and cucumber-cpp are done.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-21 14:23:50
[cucumber/common] Comment on #1507
I've raised an issue in cucumber-eclipse. It looks like renaming the branch could break their CI. <a href="https://github.com/cucumber/cucumber-eclipse/issues/433">cucumber/cucumber-eclipse#433</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-21 14:40:11
[cucumber/common] Comment on #1507
I've gone ahead and done this on cucumber-android and <a href="http://cucumber.ml">cucumber.ml</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-21 14:40:56
[cucumber/common] Comment on #1507
I think we've reached "good enough for now" on this.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-21 14:40:57
[cucumber/common] Issue closed by mattwynne
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-21 14:56:44
[cucumber/common] Comment on #249
You've made a lot of points Rien!
I appreciate the history you've been through with Cucumber-JVM since I raised this ticket back in 2017, and the clarity of thinking you can bring to this problem.
I see two sides to this problem:
1. When we compile an "empty" Gherkin scenario into a pickle
2. If/when we try to execute a pickle for an empty scenario
The original intent of this ticket was to focus on (1) - that's why I raised it in this repository, where the compiler is - because at the time empty scenarios were effectively "censored" from Cucumber by the compiler. Any debate about how to handle their execution and results would be moot because they didn't even appear in the message stream.
My feeling is that although a linter is probably a good idea to have at some point, linting is a long way away at the moment, and it still makes sense for us to think about what is the right way for the compiler to express these edge cases.
I'd like to clarify what the current behaviour of the compiler is here, as it may have changed since I raised the ticket. As <a href="https://github.com/enkessler">@enkessler</a> has pointed out, it also seems to be doing some stuff that's inconsistent with what we'd intended.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-21 15:06:23
[cucumber/common] Comment on #249
For compilation, my take on the rules would be (and I don't know what the current behaviour is yet):
[1] i.e. the background steps are only slurped up into scenarios that are non-empty)
At a common-sense/<a href="https://en.wikipedia.org/wiki/Principle_of_least_astonishment">POLS</a> level, those rules seem to make sense to me.
The one I'm not sure about is a scenario outline. If we have a scenario outline with steps but no examples, should that also be represented as a single empty pickle?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-21 15:45:06
[cucumber/common] Issue opened by mattwynne
<p><strong>Summary</strong></p>
<p>Running <code>make</code> in the latest cucumber-build container fails when using <code>arm64</code> architecture (e.g. an Apple M1)</p>
<p><strong>Expected Behavior</strong></p>
<p>The build works, as it does on CI and on a machine with <code>amd64</code> architecture.</p>
<p><strong>Current Behavior</strong></p>
<p>Java and JavaScript seem to work OK. Right now, I have hit a problem compiling the <code>gherkin</code> package for Go. There may be other problems that will appear once we're past this one however.</p>
<pre><code>ata/good/background.feature | jq --sort-keys --compact-output &gt; acceptance/testdata/good/background.feature.ast.ndjson
qemu: uncaught target signal 11 (Segmentation fault) - core dumped
diff --unified &lt;(jq "." testdata/good/background.feature.ast.ndjson) &lt;(jq "." acceptance/testdata/good/background.feature.ast.ndjson)
--- /dev/fd/63 2021-05-21 22:39:08.415226002 +0000
+++ /dev/fd/62 2021-05-21 22:39:08.415226002 +0000
@@ -1,92 +0,0 @@
-{
- "gherkinDocument": {
- "comments": [],
- "feature": {
- "children": [
- {
- "background": {
- "description": "",
- "id": "1",
- "keyword": "Background",
- "location": {
- "column": 3,
- "line": 3
- },
- "name": "a simple background",
- "steps": [
- {
- "id": "0",
- "keyword": "Given ",
- "location": {
- "column": 5,
- "line": 4
- },
- "text": "the minimalism inside a background"
- }
- ]
- }
- },
- {
- "scenario": {
- "description": "",
- "examples": [],
- "id": "3",
- "keyword": "Scenario",
- "location": {
- "column": 3,
- "line": 7
- },
- "name": "minimalistic",
- "steps": [
- {
- "id": "2",
- "keyword": "Given ",
- "location": {
- "column": 5,
- "line": 8
- },
- "text": "the minimalism"
- }
- ],
- "tags": []
- }
- },
- {
- "scenario": {
- "description": "",
- "examples": [],
- "id": "5",
- "keyword": "Scenario",
- "location": {
- "column": 3,
- "line": 10
- },
- "name": "also minimalistic",
- "steps": [
- {
- "id": "4",
- "keyword": "Given ",
- "location": {
- "column": 5,
- "line": 11
- },
- "text": "the minimalism"
- }
- ],
- "tags": []
- }
- }
- ],
- "description": "",
- "keyword": "Feature",
- "language": "en",
- "location": {
- "column": 1,
- "line": 1
- },
- "name": "Background",
- "tags": []
- },
- "uri": "testdata/good/background.feature"
- }
-}
make[2]: ****** [Makefile:58: acceptance/testdata/good/background.feature.ast.ndjson] Error 1
make[2]: ****** Deleting file 'acceptance/testdata/good/background.feature.ast.ndjson'
make[2]: Leaving directory '/app/gherkin/go'
make[1]: ****** [<a href="http://default.mk:17">default.mk:17</a>: default-go] Error 2
make[1]: Leaving directory '/app/gherkin'
make: ****** [Makefile:26: default-gherkin] Error 2
</code></pre>
<p><strong>Possible Solution</strong></p>
<p>I think the segfault is probably due to some assumptions being made about the architecture in / around <a href="https://github.com/cucumber/common/blob/main/.templates/go/default.mk">here</a> about the processor architecture.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-21 17:20:44
[cucumber/common] Comment on #249
> The one I'm not sure about is a scenario outline. If we have a scenario outline with steps but no examples, should that also be represented as a single empty pickle?
I would expect an outline with no examples to generate no pickles because an outline is merely a template from which actual scenarios are generated. An empty scenario, on the other hand, is at least a definite thing. We could debate on whether it should be a pickle with no steps or a pickle with some steps (inherited from a background) but we can say that it at least exists as a discrete test case on which a consumer might decide to take some action.
Lacking example rows, however, an outline cannot be said to be any definite thing. We could justify making a thousand pickles from it as easily as justifying just one pickle because it's all being made up from nothing anyway. Zero pickles is a reasonably safe interpretation of no examples rows.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-22 01:05:17
[cucumber/common] Comment on #1562
Please remember to label issues <a href="https://github.com/mattwynne">@mattwynne</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-22 02:46:47
[cucumber/common] Comment on #1391
Here's the relevant excerpt of <a href="https://github.com/cucumber/common/blob/main/compatibility-kit/javascript/features/markdown/markdown.md">the Markdown doc we're testing with</a>:
```
### Example: DataTable
Here is an example of what it looks like:
* Given this is a typical table:
| name | age |
| ---- | --: |
| Bill | 3 |
| Jane | 6 |
```
If we follow the same pattern as the steps with doc strings in the same document, and put the table _immediately_ after the step bullet, like this...
```
### Example: DataTable
Here is an example of what it looks like:
* Given this is a typical table:
| name | age |
| ---- | --: |
| Bill | 3 |
| Jane | 6 |
```
...the problem disappears, because the Markdown AST has the `table` as a child of the `li`, and we override the entire rendering of the `li` anyway because it's a step.
Would it be reasonable to mandate that in terms of what's valid MDG?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-22 02:52:53
[cucumber/common] Comment on #1391
Yes I think that’s reasonable. So we’ll only recognise datatables that have 2 or more spaces indentation? Need to check the spec....
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-22 02:56:29
[cucumber/common] Comment on #1391
Pushed something that deals with lists better - when we come to render a `ul`, we check if there's a step on that line (since the `ul` and its first `li` will have the same starting line in the source) - if so we render a steps list, otherwise just a regular one (with bullets).
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-22 06:00:22
[cucumber/common] Comment on #249
> For compilation, my take on the rules would be (and I don't know what the current behaviour is yet):
Might be good to do your homework first. 😛
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-22 20:52:53
[cucumber/common] Comment on #249
> > For compilation, my take on the rules would be (and I don't know what the current behaviour is yet):
>
> Might be good to do your homework first. 😛
If we take the idea of "Write the code you wish you had" up a level and make it "Write the requirements you wish you had", then not knowing the current behavior might actually be a good thing. It could free one from the temptation of altering the requirements to reduce the effort needed to implement a change rather than sticking to the desired goal and worrying about how to get there until after it is settled.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-23 14:06:13
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
TypeStrong/ts-node
<strong><a href="https://togithub.com/TypeStrong/ts-node/releases/v10.0.0"><code>v10.0.0</code></a></strong></p>
<p><a href="https://togithub.com/TypeStrong/ts-node/compare/v9.1.1...v10.0.0">Compare Source</a></p>
<p>Questions about this release? Ask in the official discussion thread: <a href="https://togithub.com/TypeStrong/ts-node/issues/1337">#1337</a></p>
<p><em>Breaking changes are prefixed with <strong>[BREAKING]</strong></em></p>
<p><strong>Added</strong></p>
<p>• Adds <code>--show-config</code> to log the resolved configuration (<a href="https://typestrong.org/ts-node/docs/troubleshooting#understanding-configuration">docs</a>) (<a href="https://togithub.com/TypeStrong/ts-node/issues/1100">#1100</a>, <a href="https://togithub.com/TypeStrong/ts-node/issues/1243">#1243</a>)
• Bundle and re-export <a href="https://togithub.com/tsconfig/node">@tsconfig/node</a><em>* configurations for convenience (<a href="https://typestrong.org/ts-node/docs/configuration#tsconfigbases">docs</a>) (<a href="https://togithub.com/TypeStrong/ts-node/issues/1202">#1202</a>, <a href="https://togithub.com/TypeStrong/ts-node/issues/1236">#1236</a>, <a href="https://togithub.com/TypeStrong/ts-node/issues/1313">#1313</a>)
• Default to appropriate <a href="https://togithub.com/tsconfig/node">@tsconfig/node</a></em>* configuration based on node and typescript versions (<a href="https://typestrong.org/ts-node/docs/configuration#default-config">docs</a>) (<a href="https://togithub.com/TypeStrong/ts-node/issues/1202">#1202</a>, <a href="https://togithub.com/TypeStrong/ts-node/issues/1236">#1236</a>, <a href="https://togithub.com/TypeStrong/ts-node/issues/1313">#1313</a>)
• Automatically reference <a href="https://togithub.com/types/node">@types/node</a>; use globally-installed <a href="https://togithub.com/types/node">@types/node</a> if not locally installed (<a href="https://togithub.com/TypeStrong/ts-node/issues/1240">#1240</a>, <a href="https://togithub.com/TypeStrong/ts-node/issues/1257">#1257</a>)
• Add <code>swc</code> integration and new <code>--transpiler</code> option to use third-party transpilers for a massive speed boost on large codebases (<a href="https://typestrong.org/ts-node/docs/transpilers">docs</a>) (<a href="https://togithub.com/TypeStrong/ts-node/issues/779">#779</a>, <a href="https://togithub.com/TypeStrong/ts-node/issues/1160">#1160</a>)
• Add <code>scopeDir</code> API option (<a href="https://typestrong.org/ts-node/api/interfaces/RegisterOptions.html#scopeDir">docs</a>) (<a href="https://togithub.com/TypeStrong/ts-node/issues/1155">#1155</a>)
• Add <code>projectSearchDir</code> API option (<a href="https://typestrong.org/ts-node/api/interfaces/RegisterOptions.html#projectSearchDir">docs</a>) (<a href="https://togithub.com/TypeStrong/ts-node/issues/1155">#1155</a>)
• Add <code>--cwd-mode</code> and <code>ts-node-cwd</code> to resolve config file relative to cwd, not entrypoint script (<a href="https://togithub.com/TypeStrong/ts-node/issues/1155">#1155</a>)</p>
<p><strong>Changed</strong></p>
<p>• <strong>[BREAKING]</strong> Make <code>--script-mode</code> default behavior; resolve tsconfig relative to entrypoint script instead of cwd (<a href="https://togithub.com/TypeStrong/ts-node/issues/949">#949</a>, <a href="https://togithub.com/TypeStrong/ts-node/issues/1197">#1197</a>, <a href="https://togithub.com/TypeStrong/ts-node/issues/1155">#1155</a>)
• In most cases this change will have no noticeable effect
• Primarily benefits portable shell scripts on your <code>$PATH</code>, because <code>ts-node</code> will respect the script's local <code>tsconfig.json</code>
• Use <code>--cwd-mode</code> or <code>ts-node-cwd</code> if you need legacy behavior
• <strong>[BREAKING]</strong> <code>ignore</code> rules evaluate relative to <code>tsconfig.json</code> directory, otherwise <code>cwd</code> (<a href="https://togithub.com/TypeStrong/ts-node/issues/1155">#1155</a>)
• <strong>[BREAKING]</strong> Remove support for node 10. Minimum supported version is node 12 (<a href="https://togithub.com/TypeStrong/ts-node/issues/1312">#1312</a>)
• Rename <code>--dir</code> to <code>--cwd</code>; rename <code>TS_NODE_DIR</code> to <code>TS_NODE_CWD</code> (<a href="https://togithub.com/TypeStrong/ts-node/issues/1155">#1155</a>)
• <code>--dir</code> and <code>TS_NODE_DIR</code> are deprecated but still parsed for backwards-compatibility
• <code>--dir</code> effectively changed the working directory of <code>ts-node</code>; renaming makes this behavior more obvious</p>
<p><strong>Deprecated</strong></p>
<p>• Deprecate <code>TS_NODE_SCOPE</code> (<a href="https://togithub.com/TypeStrong/ts-node/issues/1155">#1155</a>)
• Deprecate <code>--dir</code> and <code>TS_NODE_DIR</code> (<a href="https://togithub.com/TypeStrong/ts-node/issues/1155">#1155</a>)</p>
<p><strong>Removed</strong></p>
<p>• <strong>[BREAKING]</strong> Internal APIs removed from type declarations (<a href="https://togithub.com/TypeStrong/ts-node/issues/1242">#1242</a>)
• Removed <code>DEFAULTS</code>, <code>normalizeSlashes</code>, <code>parse</code>, <code>split</code>
• No features were removed
• This will only affect consumers of <code>ts-node</code>'s programmatic API</p>
<p><strong>Fixed</strong></p>
<p>• <strong>[BREAKING]</strong> Fix <a href="https://togithub.com/TypeStrong/ts-node/issues/1229">#1229</a> and <a href="https://togithub.com/TypeStrong/ts-node/issues/1235">#1235</a>: always throw <code>ERR_REQUIRE_ESM</code> when attempting to execute ESM as CJS, even when not using <code>--loader ts-node/esm</code> (<a href="https://togithub.com/TypeStrong/ts-node/issues/1232">#1232</a>)
• This aligns our behavior with vanilla <code>node</code>
• <strong>[BREAKING]</strong> Fix <a href="https://togithub.com/TypeStrong/ts-node/issues/1225">#1225</a>: <code>compiler</code> is loaded relative to <code>tsconfig.json</code> instead of entrypoint script (<a href="https://togithub.com/TypeStrong/ts-node/issues/1155">#1155</a>)
• In most cases this change will have no noticable effect
• Fix <a href="https://togithub.com/TypeStrong/ts-node/issues/1217">#1217</a>: REPL not always using passed stdout and stderr (<a href="https://togithub.com/TypeStrong/ts-node/issues/1224">#1224</a>)
• Fix <a href="https://togithub.com/TypeStrong/ts-node/issues/1220">#1220</a>: <code>ts-node ./index</code> may execute the wrong file extension because tsconfig search poisons the <code>require.resolve</code> cache (<a href="https://togithub.com/TypeStrong/ts-node/issues/1155">#1155</a>)
• Fix <a href="https://togithub.com/TypeStrong/ts-node/issues/1322">#1322</a>: Sourcemaps fail for filenames with spaces or other characters which are percent-encoded in URLs (<a href="https://togithub.com/TypeStrong/ts-node/issues/1160">#1160</a>, <a href="https://togithub.com/TypeStrong/ts-node/issues/1330">#1330</a>)
• Fix <a href="https://togithub.com/TypeStrong/ts-node/issues/1331">#1331</a>: Resolution of node builtin modules in ESM loader fails on node >=12.20.0, <13 (<a href="https://togithub.com/TypeStrong/ts-node/issues/1332">#1332</a>)</p>
<p><strong>Docs</strong></p>
<p>• New documentation website: <a href="https://typestrong.org/ts-node">https://typestrong.org/ts-node</a>
• README is generated to match the website
• Added page explaining CommonJS vs ESM
• Added page with Performance advice
• Added Troubleshooting page
• Organized and added to "Recipes" section with third-party tool integrations
• Added TypeDoc-generated API docs
• Work was spread across multiple tickets: <a href="https://togithub.com/TypeStrong/ts-node/issues/1207">#1207</a>, <a href="https://togithub.com/TypeStrong/ts-node/issues/1213">#1213</a>, <a href="https://togithub.com/TypeStrong/ts-node/issues/1221">#1221</a>, <a href="https://togithub.com/TypeStrong/ts-node/issues/1228">#1228</a>, <a href="https://togithub.com/TypeStrong/ts-node/issues/1244">#1244</a>, <a href="https://togithub.com/TypeStrong/ts-node/issues/1250">#1250</a>, <a href="https://togithub.com/TypeStrong/ts-node/issues/1294">#1294</a>, <a href="https://togithub.com/TypeStrong/ts-node/issues/1295">#1295</a>, <a href="https://togithub.com/TypeStrong/ts-node/issues/1296">#1296</a>, <a href="https://togithub.com/TypeStrong/ts-node/issues/1297">#1297</a>
• Thanks to these contributors for PRs which improved our documentation
• add troubleshooting tip for syntax errors (<a href="https://togithub.com/TypeStrong/ts-node/issues/1201">#1201</a>) <a href="https://togithub.com/jedwards1211">@jedwards1211</a>
• Clarify handling of tsx/jsx file extensions (<a href="https://togithub.com/TypeStrong/ts-node/issues/1179">#1179</a>) <a href="https://togithub.com/NaridaL">@NaridaL</a>
• Added <code>CONTRIBUTING.md</code> to document the codebase and our development workflow</p>
<p><a href="https://github.com/TypeStrong/ts-node/milestone/1">https://github.com/TypeStrong/ts-node/milestone/1</a></p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry t…</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-23 15:03:19
[cucumber/common] Pull request opened by aslakhellesoy
<p><strong>Summary</strong></p>
<p>Data Tables and Examples Tables will only be parsed if thet are indented 2-5 spaces. Otherwise they will be ignored.</p>
<p><strong>Details</strong></p>
<p>See discussion here: <a href="https://github.com/cucumber/common/pull/1391#discussion_r637381820">#1391 (comment)</a></p>
<p>This change also contains some changes to the CCK's <code>markdown.md</code> file so that we have a step with all of the following:</p>
<p>• Data Table
• Attachment
• Exception</p>
<p>This should make it easier to fix the double rendering issues mentioned in <a href="https://github.com/cucumber/common/pull/1391">#1391</a></p>
<p><strong>Motivation and Context</strong></p>
<p>As suggested by <a href="https://github.com/davidjgoss">@davidjgoss</a> in <a href="https://github.com/cucumber/common/pull/1391#discussion_r637381820">#1391 (comment)</a>, indentation is needed in order for Markdown parsers (such as <code>react-markdown</code>) to consider content as part of a list item.</p>
<p>Requiring indentation for Examples Tables solves another problem - distinguising between "content" tables and "Gherkin" tables. Without this distinction, a GFM table at an "unexpected" location (such as at the beginning of a document) would result in a parse error.</p>
<p>Introducing this distinction would cause unindented tables to be ignored by the Gherkin parser.</p>
<p>The reason for the 2-5 space indentation rule is that this is the range for which markdown parsers consider a table in a bullet to still be a table.</p>
<p>Tables not in a bullet will still be recognised as a table for 2-3 indents (but not 4 or more), so 2-5 seemed a reasonable range. We may want to be more restrictive, say exactly 2 space indent, or 2-3 spaces - I'm not sure...</p>
<p><strong>How Has This Been Tested?</strong></p>
<p>There are new unit tests as well as acceptance tests in Gherkin.</p>
<p><strong>Types of changes</strong></p>
<p>☑︎ Bug fix (non-breaking change which fixes an issue).
☐ New feature (non-breaking change which adds functionality).
☐ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<p><strong>Checklist:</strong></p>
<p>☐ The change has been ported to Java.
☐ The change has been ported to Ruby.
☑︎ The change has been ported to JavaScript.
☐ The change has been ported to Go.
☐ The change has been ported to .NET.
☐ I've added tests for my code.
☐ My change requires a change to the documentation.
☐ I have updated the documentation accordingly.
☐ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-23 15:12:25
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-23 15:34:15
[cucumber/common] Comment on #1391
I accidentally edited the summary in this PR thinking I was editing <a href="https://github.com/cucumber/common/pull/1564">#1564</a> - I've done my best to restore it (reverting the edits wasn't possible).
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-24 00:59:18
[cucumber/common] davidjgoss approved aslakhellesoy's pull request
LGTM
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-24 01:17:05
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-24 05:10:12
[cucumber/common] Comment on #1391
I've made some more improvements to Markdown. The Markdown rendering and AST rendering is now very similar:
Markdown:
<a href="https://user-images.githubusercontent.com/1000/119345743-388d5b80-bc91-11eb-8b2a-242058dc0fd3.png">Screenshot 2021-05-24 at 13 08 46</a>
Gherkin:
<a href="https://user-images.githubusercontent.com/1000/119345820-4e028580-bc91-11eb-8437-5d242c3e37e5.png">Screenshot 2021-05-24 at 13 09 29</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-24 06:41:58
[cucumber/common] Pull request opened by aslakhellesoy
<p>See discussion: <a href="https://github.com/cucumber/cucumber-js/pull/1645#discussion_r634999903">cucumber/cucumber-js#1645 (comment)</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-24 08:18:09
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-24 08:29:05
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
cucumber/messages-go
<strong><a href="https://togithub.com/cucumber/messages-go/compare/v16.0.0...v16.0.1"><code>v16.0.1</code></a></strong></p>
<p><a href="https://togithub.com/cucumber/messages-go/compare/v16.0.0...v16.0.1">Compare Source</a></p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-24 08:44:27
[cucumber/common] New release Release - cucumber-gherkin/v19.0.3 published by cukebot
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-24 09:24:16
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-24 11:01:41
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
protocolbuffers/protobuf
<strong><a href="https://togithub.com/protocolbuffers/protobuf/releases/v3.17.1"><code>v3.17.1</code></a></strong></p>
<p><a href="https://togithub.com/protocolbuffers/protobuf/compare/v3.17.0...v3.17.1">Compare Source</a></p>
<p><strong>PHP</strong></p>
<p>• Fixed JSON parser to allow multiple values from the same oneof as long as all but one are null.</p>
<p><strong>Ruby</strong></p>
<p>• Fixed memory bug: properly root repeated/map field when assigning. (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8639">#8639</a>)
• Fixed JSON parser to allow multiple values from the same oneof as long as all but one are null.</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-24 12:46:29
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-24 14:59:50
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
sinonjs/sinon
<strong><a href="https://togithub.com/sinonjs/sinon/blob/master/CHANGELOG.md#%E2%80%8B1100--2021-05-24|<code">https://togithub.com/sinonjs/sinon/blob/master/CHANGELOG.md#%E2%80%8B1100--2021-05-24|<code</a>v11.0.0</code>></strong></p>
<p><a href="https://togithub.com/sinonjs/sinon/compare/v10.0.1...36eaa05a98a067b0c30d128ec1c4be9420ed9436">Compare Source</a></p>
<p>==================</p>
<p>• Explicitly use samsam 6.0.2 with fix for <a href="https://togithub.com/sinonjs/sinon/issues/2345">#2345</a>
• Update most packages (<a href="https://togithub.com/sinonjs/sinon/issues/2371">#2371</a>)
• Update compatibility docs (<a href="https://togithub.com/sinonjs/sinon/issues/2366">#2366</a>)
• Update packages (includes breaking fake-timers change, see <a href="https://togithub.com/sinonjs/sinon/issues/2352">#2352</a>)
• Warn of potential memory leaks (<a href="https://togithub.com/sinonjs/sinon/issues/2357">#2357</a>)
• Fix clock test errors</p>
<p><strong><a href="https://togithub.com/sinonjs/sinon/blob/master/CHANGELOG.md#%E2%80%8B1001--2021-04-08|<code">https://togithub.com/sinonjs/sinon/blob/master/CHANGELOG.md#%E2%80%8B1001--2021-04-08|<code</a>v10.0.1</code>></strong></p>
<p><a href="https://togithub.com/sinonjs/sinon/compare/v10.0.0...v10.0.1">Compare Source</a></p>
<p>==================</p>
<p>• Upgrade sinon components (bumps y18n to 4.0.1)
• Bump y18n from 4.0.0 to 4.0.1</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-24 17:39:21
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about these updates again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-25 01:14:24
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-25 01:14:43
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-25 01:16:16
[cucumber/common] Comment on #1391
<a href="https://github.com/davidjgoss">@davidjgoss</a> do you think we can merge this to main now?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-25 02:14:46
[cucumber/common] Comment on #1391
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> yep I'm happy if you are.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-25 02:21:34
[cucumber/common] Comment on #1391
Let's do it! I'll create a new issue with various remaining styling niggles I'd like to address.
I'll merge when the CI is green.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-25 03:21:58
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-25 04:00:32
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-25 10:01:55
[cucumber/common] Comment on #1216
I'm also seeing this issue with v6.0.0 cucumber-json-formatter (on both MAC and Linux). Is there any chance to prioritize for fixing this issue?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-25 10:03:45
[cucumber/common] Comment on #1216
I'm also seeing this issue with v6.0.0 cucumber-json-formatter (on both MAC and Linux). Do we have an ETA?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-25 10:04:04
[cucumber/common] Comment on #1216
I'm also seeing this issue with v6.0.0 cucumber-json-formatter (on both MAC and Linux). Any ETA on when will this issue be fixed?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-25 13:51:17
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-25 17:04:23
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 07:25:12
[cucumber/common] Issue opened by aslakhellesoy
<p>We're currently importing CSS from the React components. That gave me an error when using this in a Next.js app (Cucumber Reports):</p>
<pre><code>14:11:04.416 | Failed to compile.
-- | --
14:11:04.416 | ../../node_modules/@cucumber/react/dist/src/components/app/GherkinDocumentList.module.scss
14:11:04.417 | CSS Modules cannot be imported from within node_modules.
14:11:04.417 | Read more: <https://nextjs.org/docs/messages/css-modules-npm>
14:11:04.417 | Location: ../../node_modules/@cucumber/react/dist/src/components/app/GherkinDocumentList.js
</code></pre>
<p>More info here: <a href="https://nextjs.org/docs/messages/css-modules-npm">https://nextjs.org/docs/messages/css-modules-npm</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 08:34:40
[cucumber/common] Comment on commit "Fix bug in json schema for TestCaseFinished"
I'm a bit confused about this one. We're already sending `testCase` messages that contain `testStep` messages. Why send them again? That's going to increase the size of the message stream considerably.
<a href="https://github.com/cucumber/common">cucumber/common</a>
matt-SmartBear
2021-05-26 08:50:20
*Thread Reply:* We're just playing with something in a branch, it's not really ready for feedback yet.
GitHub
2021-05-26 08:42:36
[cucumber/common] Comment on #1573
Got it working with <a href="https://www.npmjs.com/package/next-transpile-modules">https://www.npmjs.com/package/next-transpile-modules</a>.
Thanks <a href="https://github.com/davidjgoss">@davidjgoss</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 08:42:37
[cucumber/common] Issue closed by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 08:52:18
[cucumber/common] Comment on commit "Fix bug in json schema for TestCaseFinished"
It is specified as a required property, but was not defined. The schema was not valid.
Maybe the fix is actually to remove it from the required?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 08:52:31
[cucumber/common] Comment on commit "Fix bug in json schema for TestCaseFinished"
We were playing with a 3rd party codegen library which picked up a problem in this schema doc. The <a href="https://github.com/cucumber/common/blob/f7a9d02a326020b061d9f2b6945cb943aa2e6220/messages/jsonschema/TestCaseFinished.json#L4">list of required fields</a> at the top of the doc says that `testSteps` is required, but then we don't define it lower down. The codegen library picked up on that as a mistake, so we thought we should fix it.
Are you saying we don't need this prop, so actually the fix would be to take it out of the `required` list?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 09:22:19
[cucumber/common] Comment on commit "Fix bug in json schema for TestCaseFinished"
The TestCaseFinished message is nor supposed to contain TestCase messages at all. Cross-check against messages.proto for reference.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 11:33:34
[cucumber/common] Issue opened by mattwynne
<p><strong>Summary</strong></p>
<p>We're currently using a plain old Ruby Hash to represent message payloads. It will be simpler for consumers if we can de-serialize the messages into simple Ruby value objects, (or DTO - Data Transfer Objects) and pass those around instead.</p>
<p><strong>Expected Behavior</strong></p>
<p>API consumers should receive messages as Ruby objects instead of hashes.</p>
<p><strong>Possible Solution</strong></p>
<p>See Aslak's advice <a href="https://github.com/cucumber/cucumber-ruby/pull/1544#issuecomment-847734909">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 11:37:20
[cucumber/common] Comment on #1574
<a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> and I did some spiking on this today. We explored using <a href="https://github.com/quicktype/quicktype/">https://github.com/quicktype/quicktype/</a> to generate the Ruby classes. It looked pretty good until we tried to run it, and discovered that the generated code has a <a href="https://github.com/quicktype/quicktype/pull/1385">dependency on a very old version</a> of `Dry::Types`.
While quicktype looked like a nice idea, and might have helped us to replace the typescript and go generators too, we think that adding a dependency on a library like `Dry::Types` is too much, since many of our users will be loading Cucumber into the same process as their app, and they may be using their own version of this library.
So having explored the alternatives, it looks like building our own generator is the best solution.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 11:38:51
[cucumber/common] Comment on commit "Fix bug in json schema for TestCaseFinished"
OK, we'll just remove the prop altogether then.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 11:47:11
[cucumber/common] Pull request opened by mattwynne
<p><strong>Summary</strong></p>
<p>Playing with <a href="https://quicktype.io/">https://quicktype.io/</a> we discovered a couple of inconsistencies in the JSON schema docs</p>
<p><strong>Details</strong></p>
<p>• ParameterType required fields listed <code>id</code> twice
• ParameterType required fields listen <code>children</code> but it was not defined
• TestCaseFinished required fields listed <code>testSteps</code> but it's not needed on that message</p>
<p><strong>Motivation and Context</strong></p>
<p>Cleaning up the JSON schema so other tools could understand and use it.</p>
<p><strong>How Has This Been Tested?</strong></p>
<p>Using <code>npx quickcheck messages/jsonschema/**.json tmp.rb</code></p>
<p><strong>Types of changes</strong></p>
<p>☑︎ Bug fix (non-breaking change which fixes an issue).
☐ New feature (non-breaking change which adds functionality).
☐ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 12:51:01
[cucumber/common] Comment on #1575
This seems to have broken the generated TypeScript code (see CI failure). I assume ˋ#ˋ means self-reference? Where in the json schema spec is that documented?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 12:51:02
[cucumber/common] aslakhellesoy requested changes on mattwynne's pull request
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 12:56:06
[cucumber/common] Comment on #1575
Since generated code is added to git, you have to run ˋmakeˋ from the ˋmessagesˋ directory and commit the modified files (if any).
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 13:24:37
[cucumber/common] Comment on #1574
Good call. I also evaluated quicktype, but ended up with a custom code generator.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 13:35:36
[cucumber/common] Comment on #1575
I haven't found any mentions of `#` meaning self-reference. <a href="https://stackoverflow.com/questions/35250621/recursive-self-referencing-json-schema">This</a> SO suggests the reference needs to go via `definitions`.
And <a href="https://stackoverflow.com/questions/20752716/json-schema-recursive-schema-definition/20753545#20753545">this</a> post suggests that `#` refers to the root of the document (not "self")
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 13:45:09
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 13:45:36
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 15:02:36
[cucumber/common] Comment on #1575
Well, what _is_ the correct type for an item in the `children` array? We assumed it was instances of `ParameterType` so that `#` is meant to refer to the root of the document. Maybe `./ParameterType.json` would be better?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 15:13:27
[cucumber/common] Comment on #1575
There's some info about JSON pointers <a href="https://json-schema.org/understanding-json-schema/structuring.html#id1">here</a> and a <a href="https://datatracker.ietf.org/doc/html/rfc6901">spec</a> which talks about `#` referring to "the whole document".
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 15:52:32
[cucumber/common] Comment on #1575
> Well, what is the correct type for an item in the children array?
Sorry, my bad - I saw `children` and incorrectly assumed this:
<a href="https://github.com/cucumber/common/blob/1b64380f36cf6bb8534b59af7b064e84299c7b22/messages/messages.proto#L559-L563">common/messages/messages.proto</a>
Lines 559 to 563 in </cucumber/common/commit/1b64380f36cf6bb8534b59af7b064e84299c7b22|1b64380>
Which in JSON Schema is this:
<a href="https://github.com/cucumber/common/blob/1b64380f36cf6bb8534b59af7b064e84299c7b22/messages/jsonschema/TestCase.json#L5-L28">common/messages/jsonschema/TestCase.json</a>
Lines 5 to 28 in </cucumber/common/commit/1b64380f36cf6bb8534b59af7b064e84299c7b22|1b64380>
The `ParameterType` message doesn't have a `children` property at all:
<a href="https://github.com/cucumber/common/blob/1b64380f36cf6bb8534b59af7b064e84299c7b22/messages/messages.proto#L681-L688">common/messages/messages.proto</a>
Lines 681 to 688 in </cucumber/common/commit/1b64380f36cf6bb8534b59af7b064e84299c7b22|1b64380>
A bit of background: I used a `.proto` to `JSON Schema` conversion tool to generate the JSON schema files. The tool didn't convert everything correctly, so I've done many manual edits to the generated schemas. In the process I've apparently made several mistakes (copy paste errors) in the `required` sections. There are probably more.
These mistakes have gone unnoticed until now, because the custom Ruby code generator doesn't generate code based on the `required` array, but rather the `properties` dictionary of the schema:
• <a href="https://github.com/cucumber/common/blob/1b64380f36cf6bb8534b59af7b064e84299c7b22/messages/jsonschema/scripts/codegen.rb#L115">common/messages/jsonschema/scripts/codegen.rb</a>
Line 115 in </cucumber/common/commit/1b64380f36cf6bb8534b59af7b064e84299c7b22|1b64380>
• <a href="https://github.com/cucumber/common/blob/1b64380f36cf6bb8534b59af7b064e84299c7b22/messages/jsonschema/scripts/codegen.rb#L164">common/messages/jsonschema/scripts/codegen.rb</a>
Line 164 in </cucumber/common/commit/1b64380f36cf6bb8534b59af7b064e84299c7b22|1b64380>
This means mistakes/inconsistencies in `required` are not detected. We should probably change the code generators to iterate over `required` instead. That way we should be able to find the remaining mistakes.
Sorry again for leading you down the wrong path!
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 16:00:52
[cucumber/common] Comment on #1575
While you're working on the schemas - could you also please replace all the `"$schema": "<http://json-schema.org/draft-04/schema#>"` with `"$schema": "<https://json-schema.org/draft/2020-12/schema>`? (The conversion tool I used put those old versions in).
FWIW, there isn't currently a non-draft version of the JSON schema spec - see <a href="https://json-schema.org/specification-links.html">available versions</a>.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 16:26:42
[cucumber/common] Pull request opened by aslakhellesoy
<p>Because 4.0.0 is ESM only, and we can't (yet) use that.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 16:27:15
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-26 18:55:15
[cucumber/common] Pull request opened by renovate[bot]
<https://github.com/cucumber/common/pull/1577|#1577 Update module github.com/onsi/gomega to v1.13.0>
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
onsi/gomega
<strong><a href="https://togithub.com/onsi/gomega/releases/v1.13.0"><code>v1.13.0</code></a></strong></p>
<p><a href="https://togithub.com/onsi/gomega/compare/v1.12.0...v1.13.0">Compare Source</a></p>
<p>• Set consistently and eventually defaults on init (<a href="https://togithub.com/onsi/gomega/issues/443">#443</a>)</p>
<pre><code>Using environment variables
Closes <a href="https://togithub.com/onsi/gomega/issues/434">#434</a>
Signed-off-by: toby lorne <a href="mailto:toby@toby.codes">toby@toby.codes</a>
</code></pre>
<p>• gmeasure provides BETA support for benchmarking (<a href="https://togithub.com/onsi/gomega/issues/447">#447</a>)</p>
<pre><code>gmeasure is a new gomega subpackage intended to provide measurement and benchmarking support for durations and values. gmeasure replaces Ginkgo V1s deprecated Measure nodes and provides a migration path for users migrating to Ginkgo V2.
gmeasure is organized around an Experiment metaphor. Experiments can record several different Measurements, with each Measurement comprised of multiple data points. Measurements can hold time.Durations and float64 values and gmeasure includes support measuring the duraiton of callback functions and for sampling functions repeatedly to build an ensemble of data points. In addition, gmeasure introduces a Stopwatch abtraction for easily measuring and recording durations of code segments.
Once measured, users can readily generate Stats for Measurements to capture their key statistics and these stats can be ranked using a Ranking and associated RankingCriteria.
Experiments can be Cached to disk to speed up subsequent runs. Experiments are cached by name and version number which makes it easy to manage and bust the cache.
Finally, gmeasure integrates with Ginkgo V2 via the new ReportEntry abstraction. Experiments, Measurements, and Rankings can all be registered via AddReportEntry. Doing so generates colorful reports as part of Ginkgo's test output.
gmeasure is currently in beta and will go GA around when Ginkgo V2 goes GA.
</code></pre>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-27 00:45:33
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-27 06:18:15
[cucumber/common] Issue opened by aslakhellesoy
<p>Last year, a new <a href="https://cucumber.io/blog/bdd/gherkin-rules/">Rule</a> keyword was added to Gherkin. Please help us by adding a comment to this issue, and we'll update the parser accordingly. Example:</p>
<p>Norwegian (no): Regel</p>
<p>Obrigado!</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-27 06:47:35
[cucumber/common] Pull request merged by aslakhellesoy
<https://github.com/cucumber/common/pull/1577|#1577 Update module github.com/onsi/gomega to v1.13.0>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-27 07:14:01
[cucumber/common] Comment on #1578
Serbian (Cyrillic): Правило
Serbian (Latin): Pravilo
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-27 08:18:50
[cucumber/common] Comment on #1578
Много хвала <a href="https://github.com/pzelenovic">@pzelenovic</a>!
(In case anyone wonders - I know how to use Google Translate, but these translations should be done by humans in case we choose the wrong translation amongst many possibilities).
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-27 08:28:27
[cucumber/common] Pull request opened by l3l0
<p><strong>Details</strong></p>
<p>Update polish translation for "rule" keyword</p>
<p><strong>Motivation and Context</strong></p>
<p>Missing translation</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-27 08:58:47
[cucumber/common] Issue opened by elchupanebrej
<p><strong>Summary</strong></p>
<p>Gherkin parser doesn't interpolate parameters on background level, so hard to write tests in a DRY manner</p>
<p><strong>Current Behavior</strong></p>
<p>When I execute the next python code (gherkin-official==19.0.2)</p>
<pre><code>data = '''\
Feature: Background example
Background:
Given Use &lt;example&gt;
Rule:
Background:
Given Use &lt;example&gt;
Scenario:
Given Use &lt;example&gt;
Examples:
|example|
|car |
'''
parser = Parser()
gherkin_document = parser.parse(data)
gherkin_document['uri'] = None
pickles = Compiler().compile(gherkin_document)
pickles[0]
</code></pre>
<p>I got next pickle:</p>
<pre><code>{'astNodeIds': ['8', '6'],
'id': '3',
'name': '',
'language': 'en',
'steps': [
{'astNodeIds': ['0'], 'id': '0', 'text': 'Use &lt;example&gt;'},
{'astNodeIds': ['2'], 'id': '1', 'text': 'Use &lt;example&gt;'},
{'astNodeIds': ['4', '6'], 'id': '2', 'text': 'Use car'}
],
'tags': [],
'uri': None}
</code></pre>
<p>Pay attention that the first two steps are not interpolated</p>
<p><strong>Expected Behavior</strong></p>
<p>Both steps (Scenario and background are interpolated)</p>
<p><strong>Possible Solution</strong></p>
<p>In addition to interpolation, it's a good idea to add the possibility to put "Examples:" sections on any level (Feature/Background)</p>
<p><strong>Context & Motivation</strong></p>
<p>I have a big suite of long-running E2E highly parametrized tests. I would like to have the possibility to control the time of test run by selecting specific parametrization for cases on any test suite level. Also, I would like to have parametrization on different levels of Feature/Rule/Scenario to not blow up parameters tables<br />
I use pytest_bdd for my day-to-day activities, but it has no specific features, which this project has(and vice versa). During of integration of the current parser to that project I've found that these two projects using two different dialects of the Gherkin language</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-27 09:39:27
[cucumber/common] Comment on #1580
This is a duplicate of <a href="https://github.com/cucumber/common/issues/469">#469</a> where I have explained why this won’t work
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-27 11:01:54
[cucumber/common] Comment on #1580
Thanks for the fast responce! Yes, it's definitely duplicate.
Could we just discuss how it could work because 3 years were passed since <a href="https://github.com/cucumber/common/issues/469">#469</a>
There you said that people would mix Scenario and Scenario Outlines, but now there is no big difference between Scenario, Example, Scenario Outline, and Scenario Template. Everything from this could be used for parameterized tests.
Let assume that we already have this feature in place. It's definitely an extension of language. How it could be mixed? Let me show
```
Feature: Fruit operations
Background:
Given I have a <fruit>
Scenario:
Then I could put it to <tableware>
@small
Examples:
|fruit|tableware|
|apple|plate|
|cherry|cup|
@big
Examples:
|fruit|tableware|
|watermellon|basket|
Scenario:
Then I could wash it
Examples:
|fruit|
|apple|
|cherry|
|watermellon|
```
or even
```
Feature: Fruit operations
Background:
Given I have a <fruit>
Examples:
|fruit|
|apple|
|cherry|
|watermellon|
Scenario:
Then I could put it to <tableware>
Examples:
|tableware|
|plate|
|cup|
Scenario:
Then I could wash it
```
How it could be executed:
Every block(Background, Rule, Scenario) could have an Example table(with a selected set of parameters)
Every Route contained from Steps is parametrized. For every flow, we could get a set of parameters and a set of affected Example tables. To marshall the Route to Pickle set we do next: we make a cartesian product of every Example Table and interpolate Route steps with every variant.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-27 11:04:21
[cucumber/common] Comment on #1579
Wouldn't `Reguła` be more suitable in this context?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-27 11:07:45
[cucumber/common] Comment on #1580
Ok, what would you expect the placeholder to be for the first scenario here:
```
Feature: Fruit operations
Background:
Given I have a <fruit>
Scenario: No examples, just a regular scenario
When I go to town
Scenario:
Then I could put it to <tableware>
@small
Examples:
|fruit|tableware|
|apple|plate|
|cherry|cup|
@big
Examples:
|fruit|tableware|
|watermellon|basket|
Scenario:
Then I could wash it
Examples:
|fruit|
|apple|
|cherry|
|watermellon|
```
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-27 11:29:26
[cucumber/common] Comment on #1580
This scenario will count as parameterized and has to provide Example table with some "fruit". If it's not provided parser has to throw exception with message that Examples are expected
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-27 11:35:26
[cucumber/common] Comment on #1580
Implementing this would have significant ramifications. There are several Gherkin parser implementations to modify (Java, Ruby, TypeScript, Go, C#, Elixir, Python, Perl...)
It’s not a priority for us to undertake this massive piece of work, so I’m afraid you’d have to contribute a pull request for all those languages in order to get it accepted.
Are you up for the task? 😉
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-27 11:55:44
[cucumber/common] Comment on #1580
Not ready now) But at least I would try to implement this for python. Thanks for discussion!
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-27 13:02:59
[cucumber/common] Comment on #1578
Polish (pl): Reguła.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-28 07:24:24
[cucumber/common] Comment on #1578
Thanks <a href="https://github.com/Jaahquubel">@Jaahquubel</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-28 08:15:09
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about these updates again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-28 09:42:52
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-28 21:38:10
[cucumber/common] Issue opened by utkarssh01
<p>Issue:<br />
Test class not found in selected project</p>
<p>Steps to recreate:</p>
<ol><li>Create a cucumber feature file.</li><li>Create a Step Definitions file</li><li>Create a Runner test file using Junit</li>
</ol>
<p>But when I run the Runner file , it shows all tests are passed but when I click on any of the test it shows the error "Test class not found in selected project".<br />
<a href="https://user-images.githubusercontent.com/75351740/120058044-698cd800-c065-11eb-8a3b-1c2a4d1bec6b.png">image</a></p>
<p>For example Below is my feature file:<br />
Feature: Select Site</p>
<p>In order to Tableau select site<br />
As a user<br />
I want to choose a Tableau Site to generate reports</p>
<p>Scenario Outline: Verify if Tableau allows user to select a Site<br />
Given user navigates to Tableau Site Page<br />
When user validates the homepage title<br />
Then user will click on search box<br />
And user will senter the search string<br />
| |<br />
Then user user will select the site<br />
| |<br />
And user will succesfully reach to home page<br />
Examples:<br />
| searchSite | selectSite |<br />
| Test1 | Test1 |</p>
<p>Below is my Step Definition file for the same</p>
<p>package StepDefinitions;</p>
<p>import io.cucumber.java.en.Given;<br />
import io.cucumber.java.en.Then;</p>
<p>public class SelectSiteStep<br />
{<br />
<a href="https://github.com/given">@given</a>("^user navigates to Tableau Site Page$")<br />
public void user<em>navigates</em>to<em>tableau</em>site_page() throws Throwable{<br />
// Write code here that turns the phrase above into concrete actions<br />
System.out.println("user navigates to Tableau Site page");<br />
}</p>
<pre><code>@Then("^user will click on search box$")
public void user_will_click_on_search_box() throws Throwable{
// Write code here that turns the phrase above into concrete actions
System.out.println("user will click on search box");
}
@Then("^user will senter the search string$")
public void user_will_senter_the_search_string(String searchSite) throws Throwable{
System.out.println("user will senter the search string "+searchSite);
}
@Then("^user user will select the site$")
public void user_user_will_select_the_site(String selectSite) throws Throwable{
System.out.println("user user will select the site "+selectSite);
}
@Then("^user will succesfully reach to home page$")
public void user_will_succesfully_reach_to_home_page() throws Throwable{
// Write code here that turns the phrase above into concrete actions
}
</code></pre>
<p>}</p>
<p>Below is my Test runner file</p>
<p>package com.web.automation.base;</p>
<p>import org.junit.runner.RunWith;</p>
<p>import io.cucumber.junit.Cucumber;<br />
import io.cucumber.junit.CucumberOptions;</p>
<p><a href="https://github.com/RunWith">@RunWith</a>(Cucumber.class)<br />
@CucumberOptions<br />
(<br />
features = {"src/test/java/Features"},<br />
glue = {"StepDefinitions"},<br />
monochrome = true,<br />
plugin={"junit:reports/junit/reports.xml",<br />
"pretty:target/cucumber-pretty.text",<br />
"html:target/cucumber-html-report",<br />
"json:target/cucumber.json",<br />
"usage:target/cucumber-usage.json",}</p>
<p>)<br />
public class TestRunner<br />
{</p>
<p>}</p>
<p>Please let me know why I'm seeing that error "Test class not found in selected project".</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-29 00:24:37
[cucumber/common] Comment on #1583
Can you please surround your code with triple bacticks (<a href="https://github.github.com/gfm/#fenced-code-blocks">fenced code blocks</a>)? I find this too hard to read.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-29 02:26:53
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-29 02:29:32
[cucumber/common] Comment on #1581
We should modify this PR to use the LTS ones
<a href="https://nodejs.org/en/about/releases/">https://nodejs.org/en/about/releases/</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-29 08:01:51
[cucumber/common] Pull request opened by renovate[bot]
<https://github.com/cucumber/common/pull/1584|#1584 Update module github.com/onsi/ginkgo to v1.16.3>
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
onsi/ginkgo
<strong><a href="https://togithub.com/onsi/ginkgo/releases/v1.16.3"><code>v1.16.3</code></a></strong></p>
<p><a href="https://togithub.com/onsi/ginkgo/compare/v1.16.2...v1.16.3">Compare Source</a></p>
<p><strong>1.16.3</strong>
<strong>Features</strong></p>
<p>• Measure is now deprecated and emits a deprecation warning.</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-30 14:00:42
[cucumber/common] Comment on #1580
Ok <a href="https://github.com/elchupanebrej">@elchupanebrej</a> - the place to start would be in <a href="https://github.com/cucumber/common/blob/main/gherkin/python/gherkin/pickles/compiler.py">https://github.com/cucumber/common/blob/main/gherkin/python/gherkin/pickles/compiler.py</a>.
Let's keep this open for a couple of months. As we aim to keep all Gherkin implementations consistent, it would have to be ported to all the other languages we support.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-31 02:21:19
[cucumber/common] Pull request merged by aurelien-reeves
<https://github.com/cucumber/common/pull/1584|#1584 Update module github.com/onsi/ginkgo to v1.16.3>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-31 03:17:16
[cucumber/common] Issue opened by binarymist
<p>Upgrading from:</p>
<p>"@cucumber/cucumber": "7.0.0",<br />
"@cucumber/gherkin": "17.0.2",</p>
<p>To:</p>
<p>"@cucumber/cucumber": "7.2.1",<br />
"@cucumber/gherkin-streams": "2.0.2"</p>
<p><a href="https://www.npmjs.com/package/@cucumber/gherkin-streams">doc</a> says to require the <code>fromPaths</code> like the following:</p>
<p><code>const { fromPaths } = require('@cucumber/gherkin-streams')</code></p>
<p>When I do this <code>fromPaths</code> is <code>undefined</code>.</p>
<p>Although if I run in the debugger:</p>
<p><code>require('@cucumber/gherkin-streams');</code></p>
<p>I get:</p>
<pre><code>{GherkinStreams: {…}, __esModule: true}
GherkinStreams: {fromPaths: ƒ, fromSources: ƒ}
__esModule: true
__proto__: Object
</code></pre>
<p>Is there some sort of additional undocumented trick to require <code>fromPaths</code> or is this a bug?</p>
<p>I've read the <a href="https://github.com/cucumber/common/blob/main/gherkin/CHANGELOG.md">Changelog</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-31 04:18:17
[cucumber/common] Comment on #1585
Yes it does, thanks <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a>.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-31 05:07:15
[cucumber/common] Issue opened by binarymist
<p><strong>Summary</strong></p>
<p>In upgradeing from "@cucumber/cucumber": "7.0.0" to "@cucumber/cucumber": "7.2.1" as per <a href="https://github.com/cucumber/common/issues/1585">this issue</a> I had problems at 7.1.0.<br />
In looking at the <a href="https://github.com/cucumber/cucumber-js/blob/main/CHANGELOG.md#fixed-2">Changelog</a>, I see <a href="https://github.com/cucumber/cucumber-js/pull/1608/files">pr 1608</a> introduced a <code>stream.on</code> (in a ts file (I'm using js)). Correct me if I'm wrong, but this seems to assume that the cucumber stdout has an <code>on</code> function. I haven't seen this documented anywhere, what is the purpose of this? Also the actual change doesn't seem to be in the Changelog. I also can't find it in the <a href="https://github.com/cucumber/cucumber-js">https://github.com/cucumber/cucumber-js</a> repository. I can only see it when I npm install @cucumber/cucumber 7.1.0. Where is this code?</p>
<p>I've added <a href="https://github.com/purpleteam-labs/purpleteam-app-scanner/blob/06c5148c9f080388164a74ccf990e8ae18f7316f/src/scripts/runCuc.js#L42-L45">this</a> to get around the problem, but I have no idea what the <code>on</code> function is supposed to do?</p>
<p><a href="https://github.com/purpleteam-labs/purpleteam-app-scanner/blob/06c5148c9f080388164a74ccf990e8ae18f7316f/src/scripts/runCuc.js#L63-L67">This is where/how</a> I've been instantiating the CLI for years.</p>
<p><a href="https://github.com/purpleteam-labs/purpleteam-app-scanner/blob/06c5148c9f080388164a74ccf990e8ae18f7316f/src/api/app/models/app.parallel.js#L502-L505">This is how</a> I've been handling the CLI erros for years.</p>
<p>Is my empty <code>on</code> function the way to do this, I'm struggling to understand what's going on here?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-31 05:31:01
[cucumber/common] Comment on #1586
Some history here:
• Initially Gherkin was dependent to Node due to usage of <a href="https://nodejs.org/api/stream.html">Nodejs stream api</a>
• At some point we wanted to be able to execute Gherkin in the browser. To make that possible easily, usage of streams has been moved from the gherkin library to an independent package: gherkin-streams
`stream.on` actually uses the <a href="https://nodejs.org/api/events.html#events_emitter_on_eventname_listener">Nodejs EventEmitter api</a>.
This is just the event model of nodejs.
You say you have problem upgrading to 7.1.0. What are your problems exactly?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-31 05:43:35
[cucumber/common] Comment on #1586
The Cucumber CLI must be instanciated with a stdout which must be a <a href="https://github.com/cucumber/cucumber-js/blob/2e744ae80d84ed996e198769bb9cb33568ec5d5f/src/formatter/index.ts#L12">IFormatterStream</a>
```
export type IFormatterStream =
| FsWriteStream
| TtyWriteStream
| PassThrough
| HttpStream
```
All those are various Node streams (`fs`, `tty`, `stream`) which all are part of the node API, which uses EventEmitter API.
Try instantiating a node stream instead of a `{ ... }` object. That would be more reliable.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-31 05:57:29
[cucumber/common] Comment on #1586
> Also the actual change doesn't seem to be in the Changelog.
Actually there have been no changed in the signature of the Cucumber CLI constructor. It has always asked for a IFormatterStream, and that interface did not changed.
What have changed is the use of the `on` method from EventEmitter. But as you are using plain javascript, you did not noticed before. The CLI expect the `stdout` parameter to be a stream since a long time. The transition to typescript simply make it more explicit.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-31 07:11:14
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
puppeteer/puppeteer
<strong><a href="https://togithub.com/puppeteer/puppeteer/blob/master/CHANGELOG.md#%E2%80%8B1000-httpsgithubcompuppeteerpuppeteercomparev911v1000-2021-05-31|<code">https://togithub.com/puppeteer/puppeteer/blob/master/CHANGELOG.md#%E2%80%8B1000-httpsgithubcompuppeteerpuppeteercomparev911v1000-2021-05-31|<code</a>v10.0.0</code>></strong></p>
<p><a href="https://togithub.com/puppeteer/puppeteer/compare/v9.1.1...v10.0.0">Compare Source</a></p>
<p><strong>⚠️ BREAKING CHANGES</strong></p>
<p>• Node.js 10 is no longer supported.</p>
<p><strong>Features</strong></p>
<p>• <strong>chromium:</strong> roll to Chromium 92.0.4512.0 (r884014) (<a href="https://togithub.com/puppeteer/puppeteer/issues/7288">#7288</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/f863f4bfe015e57ea1f9fbb322f1cedee468b857">f863f4b</a>)
• <strong>requestinterception:</strong> remove cacheSafe flag (<a href="https://togithub.com/puppeteer/puppeteer/issues/7217">#7217</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/d01aa6c84a1e41f15ffed3a8d36ad26a404a7187">d01aa6c</a>)
• expose other sessions from connection (<a href="https://togithub.com/puppeteer/puppeteer/issues/6863">#6863</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/cb285a237921259eac99ade1d8b5550e068a55eb">cb285a2</a>)
• <strong>launcher:</strong> add new launcher option <code>waitForInitialPage</code> (<a href="https://togithub.com/puppeteer/puppeteer/issues/7105">#7105</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/2605309f74b43da160cda4d214016e4422bf7676">2605309</a>), closes <a href="https://togithub.com/puppeteer/puppeteer/issues/3630">#3630</a></p>
<p><strong>Bug Fixes</strong></p>
<p>• added comments for browsercontext, startCSSCoverage, and startJSCoverage. (<a href="https://togithub.com/puppeteer/puppeteer/issues/7264">#7264</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/b75039746ac6bddf1411538242b5e70b0f2e6e8a">b750397</a>)
• modified comment for method product, platform and newPage (<a href="https://togithub.com/puppeteer/puppeteer/issues/7262">#7262</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/159d2835450697dabea6f9adf6e67d158b5b8ae3">159d283</a>)
• <strong>requestinterception:</strong> fix font loading issue (<a href="https://togithub.com/puppeteer/puppeteer/issues/7060">#7060</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/c9978d20d5584c9fd2dc902e4b4ac86ed8ea5d6e">c9978d2</a>), closes <a href="https://togithub.com/puppeteer//github.com/puppeteer/puppeteer/pull/6996/issues/issuecomment-811546501">/github.com/puppeteer/puppeteer/pull/6996#issuecomment-811546501</a> <a href="https://togithub.com/puppeteer//github.com/puppeteer/puppeteer/pull/6996/issues/issuecomment-813797393">/github.com/puppeteer/puppeteer/pull/6996#issuecomment-813797393</a> <a href="https://togithub.com/puppeteer/puppeteer/issues/7038">#7038</a>
• drop support for Node.js 10 (<a href="https://togithub.com/puppeteer/puppeteer/issues/7200">#7200</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/97c9fe2520723d45a5a86da06b888ae888d400be">97c9fe2</a>), closes <a href="https://togithub.com/puppeteer/puppeteer/issues/6753">#6753</a></p>
<p><strong><a href="https://togithub.com/puppeteer/puppeteer/compare/v9.1.0...v9.1.1">9.1.1</a> (2021-05-05)</strong>
<strong>Bug Fixes</strong></p>
<p>• make targetFilter synchronous (<a href="https://togithub.com/puppeteer/puppeteer/issues/7203">#7203</a>) (<a href="https://togithub.com/puppeteer/puppeteer/commit/bcc85a0969077d122e5d8d2fb5c1061999a8ae48">bcc85a0</a>)</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-31 08:13:58
[cucumber/common] Issue closed by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-05-31 16:56:02
[cucumber/common] Issue opened by broken-bytes
<p><strong>Summary</strong>
<strong>Expected Behavior</strong>
<strong>Current Behavior</strong></p>
<p><strong>Possible Solution</strong>
<strong>Steps to Reproduce (for bugs)</strong></p>
<p><strong>Context & Motivation</strong>
<strong>Your Environment</strong></p>
<p>• Version used:
• Operating System and version:
• Link to your project:</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-01 00:01:29
[cucumber/common] Comment on #1588
We don’t currentøy have a maintainer for the ObjC implementation.
would yoube interested in picking it up? We can give you guidance and support if you need it.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-01 01:02:23
[cucumber/common] Pull request opened by davidjgoss
<p>See <a href="https://github.com/cucumber/common/issues/1585">#1585</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-01 01:15:07
[cucumber/common] aurelien-reeves approved davidjgoss's pull request
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-01 02:38:31
[cucumber/common] Pull request merged by davidjgoss
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-01 05:08:40
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-01 05:26:44
[cucumber/common] Comment on #1579
I don't know polish
But I know that Gherkin allows to have several words for a same key.
Maybe you may have both translations: "Zasada" and "Reguła"?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-01 05:45:13
[cucumber/common] Comment on #1586
Thanks for the input <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> it was very helpful.
I've used a `Writable` stream.
Implementing the stream (in my case `cucumberCliStdout`), what do I need to expect the `chunk` argument to the `write` function to be and how should I handle it. In testing (without `decodeStrings: false`) it has been a `buffer`. Will it always be the same?
Does my <a href="https://github.com/purpleteam-labs/purpleteam-app-scanner/pull/20/files">implementation</a> look correct and sufficient?
Do I need to provide any other options to the `Writable` constructor?
I take it you don't expect the stream to be in `objectMode`?
Thanks.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-01 07:09:29
[cucumber/common] Comment on #1586
I am not confident enough with streams to give you a valuable feedback.
From what I see, it looks good. But I may totally miss something, so if anyone else confident with streams could check, that would be great.
Regarding the type of `chunk`, we cannot guaranty anything. You should check its type, not do any assumption on it. Streams accept both `Buffer` and `String`, and even `any` <a href="https://nodejs.org/docs/latest-v14.x/api/stream.html#stream_writable_write_chunk_encoding_callback_1">cf. doc</a>, you should do the same.
If I understand well the decodeString option: it automatically converts strings into buffer. So if set to true (the default), you should never received chunk as string, but always as buffer. If set to false, strings remain strings, buffers remain buffers.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-01 08:58:44
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
webpack-contrib/sass-loader
<strong><a href="https://togithub.com/webpack-contrib/sass-loader/blob/master/CHANGELOG.md#%E2%80%8B1200-httpsgithubcomwebpack-contribsass-loadercomparev1111v1200-2021-06-01|<code">https://togithub.com/webpack-contrib/sass-loader/blob/master/CHANGELOG.md#%E2%80%8B1200-httpsgithubcomwebpack-contribsass-loadercomparev1111v1200-2021-06-01|<code</a>v12.0.0</code>></strong></p>
<p><a href="https://togithub.com/webpack-contrib/sass-loader/compare/v11.1.1...v12.0.0">Compare Source</a></p>
<p><strong>⚠️ BREAKING CHANGES</strong></p>
<p>• minimum supported <code>Node.js</code> version is <code>12.13.0</code></p>
<p><strong>Bug Fixes</strong></p>
<p>• crash in custom importers with worker threads (<a href="https://togithub.com/webpack-contrib/sass-loader/issues/958">#958</a>) (<a href="https://togithub.com/webpack-contrib/sass-loader/commit/67aa1391c12013aae70e08f5bbabb94e74b10a6d">67aa139</a>)
• resolving <code>_index.import.scss</code>/<code>index.import.scss</code> in packages (<a href="https://togithub.com/webpack-contrib/sass-loader/issues/906">#906</a>) (<a href="https://togithub.com/webpack-contrib/sass-loader/commit/6641a16d510db653fbdc2bcfc265603c9f6fcd1a">6641a16</a>)</p>
<p><strong><a href="https://togithub.com/webpack-contrib/sass-loader/compare/v11.1.0...v11.1.1">11.1.1</a> (2021-05-13)</strong>
<strong>Bug Fixes</strong></p>
<p>• disabled auto importing <code>fiber</code> on node >= 16 due incompatibility problems (<a href="https://togithub.com/webpack-contrib/sass-loader/issues/950">#950</a>) (<a href="https://togithub.com/webpack-contrib/sass-loader/commit/4ca004b9e733bd2261ec68a7db4de07d79cf331a">4ca004b</a>)</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-01 16:05:55
[cucumber/common] Comment on #1578
Turkish (tr): Kural
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-01 17:10:37
[cucumber/common] Comment on #1588
I would rather take the existing codebase, transfer it to swift and go from there. I only use this parser because of the abstinence of a Swift version.
I would love to help out on that though.
I was just making false assumptions. I thought generating the parser would generate all the files and features, while in fact the generated part is really just the one main handler. This got me confused because there is a new version for it(The gherkin and objc grammar files) but the codebase was outdated.
I can surely work on a swift version of that.
I do have some experience with objective c but not nearly as much as with Swift, that's why I'd prefer to transfer it over to Swift. Plus the fact that Swift is now Apple's main language.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-01 19:26:38
[cucumber/common] Comment on #1586
Hmm... So you don't know what Cucumber will be writing? It seems strange that you can't say what it will be passing to `write`. If the Cucumber team doesn't know, what hope do consumers have with knowing how to handle the possible types?
I've been handling everything in the past as `string`. Everything (`chunk`) that comes through `write` gets sent to a `publisher`. This then gets logged to varius places including published to Redis channels.
It'd be safe to say that we've never noticed anything being received other than `string`s. Any suggestions on what the other forms would be for? Are they safe to convert to `string`? If they are not what do you suggest we do with them?
Thanks.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-02 00:35:29
[cucumber/common] Comment on #1586
> Hmm... So you don't know what Cucumber will be writing? It seems strange that you can't say what it will be passing to `write`. If the Cucumber team doesn't know, what hope do consumers have with knowing how to handle the possible types?
I did not say that. I just said that we cannot guaranty the type of the data here as this is not specific to Cucumber. This is based on a Node API which gives any guaranty on the type of the incoming data.
That stream is passed to formatters, including custom formatters. We do not know what is going on in those custom formatters.
> I've been handling everything in the past as `string`. Everything (`chunk`) that comes through `write` gets sent to a `publisher`. This then gets logged to varius places including published to Redis channels.
>
> It'd be safe to say that we've never noticed anything being received other than `string`s. Any suggestions on what the other forms would be for? Are they safe to convert to `string`? If they are not what do you suggest we do with them?
That stream representing stdout, it is relatively safe to assume they'll receive strings or data easily convertible as strings.
But as it is passed to custom formatters and end users being able to easily implement their own custom formatters, you should not trust the data and implement proper checks and guards before publishing it to your system, or before processing it by your own services.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-02 10:15:22
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-02 12:37:12
[cucumber/common] Pull request opened by renovate[bot]
<https://github.com/cucumber/common/pull/1591|#1591 Update module github.com/onsi/ginkgo to v1.16.4>
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
onsi/ginkgo
<strong><a href="https://togithub.com/onsi/ginkgo/releases/v1.16.4"><code>v1.16.4</code></a></strong></p>
<p><a href="https://togithub.com/onsi/ginkgo/compare/v1.16.3...v1.16.4">Compare Source</a></p>
<p><strong>1.16.4</strong>
<strong>Fixes</strong></p>
<p>1.16.4 retracts 1.16.3. There are no code changes. The 1.16.3 tag was associated with the wrong commit and an attempt to change it after-the-fact has proven problematic. 1.16.4 retracts 1.16.3 in Ginkgo's go.mod and creates a new, correctly tagged, release.</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-03 05:04:13
[cucumber/common] Issue opened by davidjgoss
<p>Current:</p>
<p><a href="https://user-images.githubusercontent.com/3192745/120641691-ce806d80-c46b-11eb-97b3-40be49960efb.png">image</a></p>
<p><a href="https://reports.cucumber.io/">https://reports.cucumber.io/</a></p>
<p><a href="https://user-images.githubusercontent.com/3192745/120641721-db04c600-c46b-11eb-99f8-3055ba0094b3.png">image</a></p>
<p>Will need to:</p>
<p>☐ Get <a href="https://github.com/cucumber/common/pull/1302">#1302</a> updated and merged first
☐ Add <a href="https://tailwindcss.com/">tailwindcss</a> to the tooling
☐ Copy styling from Cucumber reports for results and meta (should be easy as it's all about classes in the DOM)
☐ Come up with some good styling for the search/filter bit.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-03 06:34:22
[cucumber/common] Pull request merged by aurelien-reeves
<https://github.com/cucumber/common/pull/1591|#1591 Update module github.com/onsi/ginkgo to v1.16.4>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-03 08:41:28
[cucumber/common] Pull request opened by luke-hill
<p><strong>Summary</strong></p>
<p>Fix tagging of Rules</p>
<p><strong>Details</strong></p>
<p>During the Gherkin Query cache step where it finds all data and relates them to specific AST Node IDs, it wasn't storing tag data. This was likely because similar to features, it needs to look "up" and not down through descendents</p>
<p><strong>Motivation and Context</strong></p>
<p>Fix Ruby</p>
<p><strong>How Has This Been Tested?</strong></p>
<p>CI<br />
Altered tests in cucumber-ruby<br />
Discussion during general meeting about executing portions of the regular CCK against all flavours of cucumber</p>
<p><strong>Screenshots (if appropriate):</strong>
<strong>Types of changes</strong></p>
<p>☑︎ Bug fix (non-breaking change which fixes an issue).
☐ New feature (non-breaking change which adds functionality).
☐ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<p><strong>Checklist:</strong></p>
<p>☐ The change has been ported to Java.
☐ The change has been ported to Ruby.
☐ The change has been ported to JavaScript.
☐ The change has been ported to Go.
☐ The change has been ported to .NET.
☐ I've added tests for my code.
☐ My change requires a change to the documentation.
☐ I have updated the documentation accordingly.
☐ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-03 08:42:29
[cucumber/common] Comment on #1593
Copied this from the method above. I'm not sure this can happen (Same for features), but I'm not really sure how the parser generates the original AST which is then iterated through when we click "Run scenario" so I'll just copy and let it all work.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-03 08:49:52
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
mockito/mockito
<strong><a href="https://togithub.com/mockito/mockito/releases/v3.11.0"><code>v3.11.0</code></a></strong></p>
<p><a href="https://togithub.com/mockito/mockito/compare/v3.10.0...v3.11.0">Compare Source</a></p>
<p><em>Changelog generated by <a href="https://togithub.com/shipkit/shipkit-changelog">Shipkit Changelog Gradle Plugin</a></em></p>
<p><strong>3.11.0</strong></p>
<p>• 2021-06-03 - <a href="https://togithub.com/mockito/mockito/compare/v3.10.0...v3.11.0">18 commit(s)</a> by Charles Munger, Szczepan Faber, dependabot[bot]
• Bump versions.bytebuddy from 1.11.0 to 1.11.1 <a href="https://togithub.com/mockito/mockito/pull/2313">(#2313)</a>
• Undo parent for MultipleParentsClassLoader <a href="https://togithub.com/mockito/mockito/pull/2312">(#2312)</a>
• Bump shipkit-auto-version from 1.1.14 to 1.1.16 <a href="https://togithub.com/mockito/mockito/pull/2310">(#2310)</a>
• Bump gradle/wrapper-validation-action from 1.0.3 to 1.0.4 <a href="https://togithub.com/mockito/mockito/pull/2309">(#2309)</a>
• Bump com.github.ben-manes.versions from 0.38.0 to 0.39.0 <a href="https://togithub.com/mockito/mockito/pull/2308">(#2308)</a>
• Bump shipkit-auto-version from 1.1.11 to 1.1.14 <a href="https://togithub.com/mockito/mockito/pull/2307">(#2307)</a>
• Use the parent classloader if the context classloader is a child of it. <a href="https://togithub.com/mockito/mockito/pull/2306">(#2306)</a>
• Bump kotlin-stdlib from 1.5.0 to 1.5.10 <a href="https://togithub.com/mockito/mockito/pull/2305">(#2305)</a>
• "The type is not public and its mock class is loaded by a different class loader" with a context classloader that delegates <a href="https://togithub.com/mockito/mockito/issues/2303">(#2303)</a>
• Enabled automated changelog <a href="https://togithub.com/mockito/mockito/pull/2301">(#2301)</a>
• Bump kotlinx-coroutines-core from 1.4.3-native-mt to 1.5.0-native-mt <a href="https://togithub.com/mockito/mockito/pull/2299">(#2299)</a>
• Bump versions.errorprone from 2.6.0 to 2.7.1 <a href="https://togithub.com/mockito/mockito/pull/2298">(#2298)</a>
• Bump junit-platform-launcher from 1.7.1 to 1.7.2 <a href="https://togithub.com/mockito/mockito/pull/2297">(#2297)</a>
• Bump versions.junitJupiter from 5.7.1 to 5.7.2 <a href="https://togithub.com/mockito/mockito/pull/2296">(#2296)</a>
• Renamed main dev branch <a href="https://togithub.com/mockito/mockito/pull/2295">(#2295)</a>
• Bump gradle/wrapper-validation-action from 1 to 1.0.3 <a href="https://togithub.com/mockito/mockito/pull/2294">(#2294)</a>
• Bump actions/checkout from 2 to 2.3.4 <a href="https://togithub.com/mockito/mockito/pull/2293">(#2293)</a>
• 'this' is not available - when enabling mock-maker-inline <a href="https://togithub.com/mockito/mockito/issues/2082">(#2082)</a></p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-04 00:20:52
[cucumber/common] Comment on #1593
It looks safer that way.
Good to me, even if useless. It may be useless for now, but that can change.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-04 00:23:28
[cucumber/common] aurelien-reeves commented on luke-hill's pull request
Would it be possible to add at east one test in query_spec which would illustrate the fix?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-04 01:57:54
[cucumber/common] aurelien-reeves approved luke-hill's pull request
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-04 04:13:02
[cucumber/common] Pull request merged by luke-hill
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-04 04:57:14
[cucumber/common] Pull request merged by mpkorstanje
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-06 13:22:16
[cucumber/common] Issue opened by sohamghoshjobprofiles
<p>the <em>keyword</em> <code>rule</code> translates to <code>Rule</code> in <strong><em>Hindi</em></strong>. It should translate to <code>नियम</code> instead keeping with the pattern.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-07 06:52:36
[cucumber/common] Comment on #1575
We should generate this list automatically. Fix coming....
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-07 07:32:43
[cucumber/common] Comment on #1575
I think this should be done in a separate job - treating the jsonschemas directory as a separate module
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-07 09:06:42
[cucumber/common] Comment on #1575
<a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> <a href="https://github.com/mattwynne">@mattwynne</a> I have added a validation utility that validates all of the CCK messages against the schema. It's currently disabled, but if you comment out the next line you can see it fail.
To validate a specific file, run:
```
make acceptance/minimal/minimal.feature.ndjson.valid
```
I think we should make further changes to the schemas to make these validation checks pass. Can you two have a look at that?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-07 09:06:43
[cucumber/common] aslakhellesoy requested changes on mattwynne's pull request
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-07 09:15:11
[cucumber/common] Comment on #1588
Hi <a href="https://github.com/broken-bytes">@broken-bytes</a> - thanks for offering to help!
It's only the parser that is generated from the grammar. The lexer/scanner and AST builder logic is hand-written.
I realise it might be a bit daunting to become familiar with the ObjC codebase and the build system, but we're here to help. We usually hang out in <a href="https://cucumberbdd-slack-invite.herokuapp.com/">Slack</a> and we also have weekly committer meetings on Zoom every Thursday at 3pm BST - see the `#committers` channel for more details.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-07 10:09:13
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
mochajs/mocha
<strong><a href="https://togithub.com/mochajs/mocha/blob/master/CHANGELOG.md#%E2%80%8B900--2021-06-07|<code">https://togithub.com/mochajs/mocha/blob/master/CHANGELOG.md#%E2%80%8B900--2021-06-07|<code</a>v9.0.0</code>></strong></p>
<p><a href="https://togithub.com/mochajs/mocha/compare/v8.4.0...v9.0.0">Compare Source</a></p>
<p><strong>💥 Breaking Changes</strong></p>
<p>• <a href="https://togithub.com/mochajs/mocha/issues/4633">#4633</a>: <strong>Drop Node.js v10.x support</strong> (<a href="https://togithub.com/juergba"><strong>@juergba</strong></a>)
• <a href="https://togithub.com/mochajs/mocha/issues/4635">#4635</a>: <code>import</code>-first loading of test files (<a href="https://togithub.com/giltayar"><strong>@giltayar</strong></a>)</p>
<p><strong>Mocha is going ESM-first!</strong> This means that it will now use ESM <code>import(test_file)</code> to load the test files, instead of the CommonJS <code>require(test_file)</code>. This is not a problem, as <code>import</code> can also load most files that <code>require</code> does. In the rare cases where this fails, it will fallback to <code>require(...)</code>. This ESM-first approach is the next step in Mocha's ESM migration, and allows ESM loaders to load and transform the test file.</p>
<p>• <a href="https://togithub.com/mochajs/mocha/issues/4636">#4636</a>: Remove deprecated <code>utils.lookupFiles()</code> (<a href="https://togithub.com/juergba"><strong>@juergba</strong></a>)
• <a href="https://togithub.com/mochajs/mocha/issues/4638">#4638</a>: Limit the size of <code>actual</code>/<code>expected</code> for <code>diff</code> generation (<a href="https://togithub.com/juergba"><strong>@juergba</strong></a>)
• <a href="https://togithub.com/mochajs/mocha/issues/4389">#4389</a>: Refactoring: Consuming log-symbols alternate to code for win32 in reporters/base (<a href="https://togithub.com/MoonSupport"><strong>@MoonSupport</strong></a>)</p>
<p><strong>🎉 Enhancements</strong></p>
<p>• <a href="https://togithub.com/mochajs/mocha/issues/4640">#4640</a>: Add new option <code>--dry-run</code> (<a href="https://togithub.com/juergba"><strong>@juergba</strong></a>)</p>
<p><strong>🐛 Fixes</strong></p>
<p>• <a href="https://togithub.com/mochajs/mocha/issues/4128">#4128</a>: Fix: control stringification of error message (<a href="https://togithub.com/syeutyu"><strong>@syeutyu</strong></a>)</p>
<p><strong>🔩 Other</strong></p>
<p>• <a href="https://togithub.com/mochajs/mocha/issues/4646">#4646</a>: Deprecate <code>Runner(suite: Suite, delay: boolean)</code> signature (<a href="https://togithub.com/juergba"><strong>@juergba</strong></a>)
• <a href="https://togithub.com/mochajs/mocha/issues/4643">#4643</a>: Update production dependencies (<a href="https://togithub.com/juergba"><strong>@juergba</strong></a>)</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-07 10:21:37
[cucumber/common] Comment on #1575
I've fixed this
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-07 10:24:39
[cucumber/common] Comment on #1575
I realised the order is significant, so a fully automated list using `find` won't work until ajv fixes <a href="https://github.com/ajv-validator/ajv-cli/issues/43">ajv-validator/ajv-cli#43</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-07 10:25:02
[cucumber/common] Comment on #1575
It was an easy enough fix - done!
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-07 14:22:38
[cucumber/common] Comment on #1575
I've also added a new Markdown code generator, so our `messages/messages.md` is updated automatically.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-07 14:48:37
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-08 05:10:27
[cucumber/common] Pull request opened by aurelien-reeves
<p><strong>Summary</strong></p>
<p>Add a code generator for messages to generate ruby DTOs instead of relying on plain ruby hashes</p>
<p>Fixes <a href="https://github.com/cucumber/common/issues/1574">#1574</a></p>
<p><strong>Details</strong></p>
<p>Add templates and class to Codegen for ruby DTOs</p>
<p><strong>Motivation and Context</strong></p>
<p>Cf. <a href="https://github.com/cucumber/common/issues/1574">#1574</a></p>
<p><strong>How Has This Been Tested?</strong></p>
<p>N/A</p>
<p><strong>Types of changes</strong></p>
<p>• New feature (non-breaking change which adds functionality).</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-08 06:49:15
[cucumber/common] Comment on commit "Generate ruby DTOs"
With so many arguments, I think positional arguments become difficult to use for client code.
How about changing to <a href="https://thoughtbot.com/blog/ruby-2-keyword-arguments">keyword arguments</a>?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-08 06:49:50
[cucumber/common] Comment on commit "Generate ruby DTOs"
Ah read too fast. You _are_ using keyword arguments. 😅
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-08 06:50:06
[cucumber/common] Comment on commit "Generate ruby DTOs"
This looks great!
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-08 06:52:21
[cucumber/common] Comment on commit "Generate ruby DTOs"
😁
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-08 06:56:17
[cucumber/common] Comment on #1597
Before going further - like using those DTOs in cucumber-gherkin - I suggest to review and merge that PR as-is while it is still relatively small and easy to review.
Then we could open a new PR dedicated to using those DTOs into Gherkin.
WDYT?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-08 08:41:33
[cucumber/common] Comment on #1597
I think this is for a separate PR, but the idea occurs to me that we could include some small amount of RDoc in the generated code, perhaps just referencing some other place where we document the message protocol.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-08 09:11:51
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
protocolbuffers/protobuf
<strong><a href="https://togithub.com/protocolbuffers/protobuf/releases/v3.17.3"><code>v3.17.3</code></a></strong></p>
<p><a href="https://togithub.com/protocolbuffers/protobuf/compare/v3.17.2...v3.17.3">Compare Source</a></p>
<p><strong>C++</strong></p>
<p>• Introduce FieldAccessListener.
• Stop emitting boilerplate {Copy/Merge}From in each ProtoBuf class
• Fixed some uninitialized variable warnings in <a href="http://generated_message_reflection.cc">generated<em>message</em>reflection.cc</a>.</p>
<p><strong>Kotlin</strong></p>
<p>• Fix duplicate proto files error (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8699">#8699</a>)</p>
<p><strong>Java</strong></p>
<p>• Fixed parser to check that we are at a proper limit when a sub-message has<br />
finished parsing.</p>
<p><strong>General</strong></p>
<p>• Support M1 (<a href="https://togithub.com/protocolbuffers/protobuf/issues/8557">#8557</a>)</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-08 09:12:15
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-08 10:39:09
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about these updates again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-08 10:49:34
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-08 10:53:05
[cucumber/common] Comment on #1597
Might as well put that in now I think. The URL to the documentation is <a href="https://github.com/cucumber/common/tree/main/messages#readme">https://github.com/cucumber/common/tree/main/messages#readme</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-08 14:11:48
[cucumber/common] Comment on #1532
Mono is removed from <a href="https://github.com/cucumber/build/blob/release/CHANGELOG.md#051">version 5.0.1 of the build environment container image</a>, so all that's needed now is to change:
• <a href="https://github.com/cucumber/common/blob/main/.circleci/config.yml#L15">the build image version for CircleCI</a> and
• <a href="https://github.com/cucumber/common/blob/main/Makefile#L2">the build image version in the Makefile</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-08 22:22:52
[cucumber/common] Comment on #1562
I have a new error today, it's been a week or two since I tried to run `make` in the monorepo:
```
cukebot@186310f5f462:/app$ make
cd messages && make default
make[1]: Entering directory '/app/messages'
if [[ -d jsonschema ]]; then cd jsonschema && make default; fi
make[2]: Entering directory '/app/messages/jsonschema'
make[2]: Nothing to be done for 'default'.
make[2]: Leaving directory '/app/messages/jsonschema'
if [[ -d javascript ]]; then cd javascript && make default; fi
make[2]: Entering directory '/app/messages/javascript'
make[2]: Nothing to be done for 'default'.
make[2]: Leaving directory '/app/messages/javascript'
if [[ -d go ]]; then cd go && make default; fi
make[2]: Entering directory '/app/messages/go'
make[2]: Nothing to be done for 'default'.
make[2]: Leaving directory '/app/messages/go'
if [[ -d java ]]; then cd java && make default; fi
make[2]: Entering directory '/app/messages/java'
make[2]: Nothing to be done for 'default'.
make[2]: Leaving directory '/app/messages/java'
if [[ -d ruby ]]; then cd ruby && make default; fi
make[2]: Entering directory '/app/messages/ruby'
make[2]: Nothing to be done for 'default'.
make[2]: Leaving directory '/app/messages/ruby'
if [[ -d python ]]; then cd python && make default; fi
if [[ -d dotnet ]]; then cd dotnet && make default; fi
make[2]: Entering directory '/app/messages/dotnet'
dotnet build -c Release -p:SnapshotSuffix=""
Microsoft (R) Build Engine version 16.9.0+57a23d249 for .NET
Copyright (C) Microsoft Corporation. All rights reserved.
Determining projects to restore...
All projects are up-to-date for restore.
/home/cukebot/.nuget/packages/grpc.tools/2.38.0/build/_protobuf/Google.Protobuf.Tools.targets(280,5): error MSB6006: "/home/cukebot/.nuget/packages/grpc.tools/2.38.0/tools/linux_arm64/protoc" exited with code 139. [/app/messages/dotnet/Cucumber.Messages/Cucumber.Messages.csproj]
/home/cukebot/.nuget/packages/grpc.tools/2.38.0/build/_protobuf/Google.Protobuf.Tools.targets(280,5): error MSB6006: "/home/cukebot/.nuget/packages/grpc.tools/2.38.0/tools/linux_arm64/protoc" exited with code 139. [/app/messages/dotnet/Cucumber.Messages/Cucumber.Messages.csproj]
Build FAILED.
/home/cukebot/.nuget/packages/grpc.tools/2.38.0/build/_protobuf/Google.Protobuf.Tools.targets(280,5): error MSB6006: "/home/cukebot/.nuget/packages/grpc.tools/2.38.0/tools/linux_arm64/protoc" exited with code 139. [/app/messages/dotnet/Cucumber.Messages/Cucumber.Messages.csproj]
/home/cukebot/.nuget/packages/grpc.tools/2.38.0/build/_protobuf/Google.Protobuf.Tools.targets(280,5): error MSB6006: "/home/cukebot/.nuget/packages/grpc.tools/2.38.0/tools/linux_arm64/protoc" exited with code 139. [/app/messages/dotnet/Cucumber.Messages/Cucumber.Messages.csproj]
0 Warning(s)
2 Error(s)
Time Elapsed 00:00:05.05
make[2]: *** [<a href="http://default.mk:33">default.mk:33</a>: .built] Error 1
make[2]: Leaving directory '/app/messages/dotnet'
make[1]: *** [<a href="http://default.mk:17">default.mk:17</a>: default-dotnet] Error 2
make[1]: Leaving directory '/app/messages'
make: *** [Makefile:26: default-messages] Error 2
```
<a href="https://github.com/gasparnagy">@gasparnagy</a> <a href="https://github.com/ciaranmcnulty">@ciaranmcnulty</a> any ideas what's up? <a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> I am surprised we're running anything to do with protobuf anymore...?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-08 22:36:53
[cucumber/common] Comment on #1562
The dotnet version of messages is bit updated yet to remove protobuf. I think we can just temporarily ignore this step for now if it causes problems with arm64... Az Android Outlook<<a href="https://aka.ms/ghei36">https://aka.ms/ghei36</a>> letöltése
<#|…>
________________________________ From: Matt Wynne ***@***.***> Sent: Wednesday, June 9, 2021 7:23:04 AM To: cucumber/common ***@***.***> Cc: Gáspár Nagy ***@***.***>; Mention ***@***.***> Subject: Re: [cucumber/common] Fix the build on arm64 architecture (<a href="https://github.com/cucumber/common/issues/1562">#1562</a>) I have a new error today, it's been a week or two since I tried to run make in the monorepo: ***@***.***:/app$ make cd messages && make default make[1]: Entering directory '/app/messages' if [[ -d jsonschema ]]; then cd jsonschema && make default; fi make[2]: Entering directory '/app/messages/jsonschema' make[2]: Nothing to be done for 'default'. make[2]: Leaving directory '/app/messages/jsonschema' if [[ -d javascript ]]; then cd javascript && make default; fi make[2]: Entering directory '/app/messages/javascript' make[2]: Nothing to be done for 'default'. make[2]: Leaving directory '/app/messages/javascript' if [[ -d go ]]; then cd go && make default; fi make[2]: Entering directory '/app/messages/go' make[2]: Nothing to be done for 'default'. make[2]: Leaving directory '/app/messages/go' if [[ -d java ]]; then cd java && make default; fi make[2]: Entering directory '/app/messages/java' make[2]: Nothing to be done for 'default'. make[2]: Leaving directory '/app/messages/java' if [[ -d ruby ]]; then cd ruby && make default; fi make[2]: Entering directory '/app/messages/ruby' make[2]: Nothing to be done for 'default'. make[2]: Leaving directory '/app/messages/ruby' if [[ -d python ]]; then cd python && make default; fi if [[ -d dotnet ]]; then cd dotnet && make default; fi make[2]: Entering directory '/app/messages/dotnet' dotnet build -c Release -p:SnapshotSuffix="" Microsoft (R) Build Engine version 16.9.0+57a23d249 for .NET Copyright (C) Microsoft Corporation. All rights reserved. Determining projects to restore... All projects are up-to-date for restore. /home/cukebot/.nuget/packages/grpc.tools/2.38.0/build/_protobuf/Google.Protobuf.Tools.targets(280,5): error MSB6006: "/home/cukebot/.nuget/packages/grpc.tools/2.38.0/tools/linux_arm64/protoc" exited with code 139. [/app/messages/dotnet/Cucumber.Messages/Cucumber.Messages.csproj] /home/cukebot/.nuget/packages/grpc.tools/2.38.0/build/_protobuf/Google.Protobuf.Tools.targets(280,5): error MSB6006: "/home/cukebot/.nuget/packages/grpc.tools/2.38.0/tools/linux_arm64/protoc" exited with code 139. [/app/messages/dotnet/Cucumber.Messages/Cucumber.Messages.csproj] Build FAILED. /home/cukebot/.nuget/packages/grpc.tools/2.38.0/build/_protobuf/Google.Protobuf.Tools.targets(280,5): error MSB6006: "/home/cukebot/.nuget/packages/grpc.tools/2.38.0/tools/linux_arm64/protoc" exited with code 139. [/app/messages/dotnet/Cucumber.Messages/Cucumber.Messages.csproj] /home/cukebot/.nuget/packages/grpc.tools/2.38.0/build/_protobuf/Google.Protobuf.Tools.targets(280,5): error MSB6006: "/home/cukebot/.nuget/packages/grpc.tools/2.38.0/tools/linux_arm64/protoc" exited with code 139. [/app/messages/dotnet/Cucumber.Messages/Cucumber.Messages.csproj] 0 Warning(s) 2 Error(s) Time Elapsed 00:00:05.05 make[2]: *** [<a href="http://default.mk:33">default.mk:33</a>: .built] Error 1 make[2]: Leaving directory '/app/messages/dotnet' make[1]: *** [<a href="http://default.mk:17">default.mk:17</a>: default-dotnet] Error 2 make[1]: Leaving directory '/app/messages' make: *** [Makefile:26: default-messages] Error 2 <a href="https://github.com/gasparnagy">@gasparnagy</a><<a href="https://github.com/gasparnagy">https://github.com/gasparnagy</a>> <a href="https://github.com/ciaranmcnulty">@ciaranmcnulty</a><<a href="https://github.com/ciaranmcnulty">https://github.com/ciaranmcnulty</a>> any ideas what's up? <a href="https://github.com/aslakhellesoy">@aslakhellesoy</a><<a href="https://github.com/aslakhellesoy">https://github.com/aslakhellesoy</a>> I am surprised we're running anything to do with protobuf anymore...? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<<a href="https://github.com/cucumber/common/issues/1562#issuecomment-857384361">#1562 (comment)</a>>, or unsubscribe<<a href="https://github.com/notifications/unsubscribe-auth/AABDKPKWQGA3ENWSJHXUABLTR33DRANCNFSM45J7CH6Q">https://github.com/notifications/unsubscribe-auth/AABDKPKWQGA3ENWSJHXUABLTR33DRANCNFSM45J7CH6Q</a>>.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-08 22:52:04
[cucumber/common] mattwynne approved aurelien-reeves's pull request
<a href="https://camo.githubusercontent.com/841a0e86bac70a82044d6345f329258215d58285fa49d6633f92c577baf6dc53/68747470733a2f2f6d65646961302e67697068792e636f6d2f6d656469612f624b6e456e6436357a717866712f67697068792e676966">https://camo.githubusercontent.com/841a0e86bac70a82044d6345f329258215d58285fa49d6633f92c577baf6dc53/68747470733a2f2f6d65646961302e67697068792e636f6d2f6d656469612f624b6e456e6436357a717866712f67697068792e676966</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-08 23:49:31
[cucumber/common] Comment on #1597
Great idea!
I have even added the documentation that is already in the json schemas into the generated DTOs.
We may do that for typescript and go.
Some formatting is part of the schemas. We may also clean those a little bit.
But that can be done in other PR, I will merge this one soon to move on with using the ruby DTOs within gherkin.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 00:42:10
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 01:07:52
[cucumber/common] Pull request opened by aurelien-reeves
<p>The point of the ruby DTOs was to stick with ruby standard.<br />
We missed to convert cameCase attribute names to snake_case in the previous PR: <a href="https://github.com/cucumber/common/issues/1574">#1574</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 01:33:15
[cucumber/common] Issue opened by aslakhellesoy
<p><strong>Summary</strong></p>
<p>A Cucumber Reports used <code>@cucumber/json-to-messages</code> to convert Behave JSON to messages. One of the messages looks like this (sensitive details edited):</p>
<pre><code>{
"testStepFinished": {
"testCaseStartedId": "522e2d20-37f0-4296-87fe-b87069afccec",
"testStepId": "9c3b9582-d6d2-4f55-88e7-a6d563c57df3",
"testStepResult": {
"willBeRetried": false,
"duration": {
"seconds": 0,
"nanos": 0
},
"status": "FAILED",
"message": [
"Assertion Failed: The pattern isn't contained!!!",
"Captured stdout:",
"",
"",
"login:",
" admin",
"Password:",
" ",
" Bla bla bla bla ",
"",
"--------------------------------------------------------------------------------",
"",
"Bla bla bla,
" bla bla"
]
},
"timestamp": {
"seconds": 1623180174,
"nanos": 354000000
}
}
}
</code></pre>
<p><strong>Expected Behavior</strong></p>
<p>The <code>message</code> field should be a string (with newlines)</p>
<p><strong>Current Behavior</strong></p>
<p>The <code>message</code> field is a string array</p>
<p><strong>Possible Solution</strong></p>
<p>Analyse the output from Behave and determine if the JSON output contains string arrays. If it does, join the string array with <code>\n</code> before setting the <code>message</code> field.</p>
<p>We should also perform JSON schema validation of the generated messages in the same way as we now do for the CCK.<br />
See</p>
<p><a href="https://github.com/cucumber/common/blob/9bf5f7638f0ac4eb15d2345e728694d9a755bd0b/compatibility-kit/javascript/Makefile#L20-L22">common/compatibility-kit/javascript/Makefile</a></p>
<p>Lines 20 to 22 in </cucumber/common/commit/9bf5f7638f0ac4eb15d2345e728694d9a755bd0b|9bf5f76></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 02:12:36
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 02:12:47
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 02:14:48
[cucumber/common] Comment on #1304
I encounter the same problem with version 6.10.4. Is there a workaround until there is a fix?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 03:07:24
[cucumber/common] Pull request opened by aurelien-reeves
<p>Requires <a href="https://github.com/cucumber/common/pull/1601">#1601</a> to be merged first</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 05:25:07
[cucumber/common] Pull request opened by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 05:26:00
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 05:45:37
[cucumber/common] Comment on #1603
<a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> after merging this, can we cut a new version of gherkin so I can then get my PR updated in the regular cucumber-ruby
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 06:29:06
[cucumber/common] Pull request merged by mattwynne
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 06:30:54
[cucumber/common] Comment on #1302
What's left to do to get this merged? <a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> <a href="https://github.com/davidjgoss">@davidjgoss</a> do you still want to review it?
Looks like it will need a rebase now.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 06:41:29
[cucumber/common] Comment on #1603
> <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> after merging this, can we cut a new version of gherkin so I can then get my PR updated in the regular cucumber-ruby
That's the plan
But it is not finished yet.
We have to write a json serializer for those DTOs
And to check the compatibility of all the modules with those new messages and gherkin.
But it is going well at the moment :)
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 06:42:24
[cucumber/common] Comment on #1603
Isn't `JSON.stringify(envelope)` enough?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 06:44:36
[cucumber/common] Comment on #1603
> Isn't `JSON.stringify(envelope)` enough?
This is not javascript, this is ruby
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 06:45:41
[cucumber/common] Comment on #1601
It seems to me that converting to `snake_case` complicates everything a lot. If we weren't doing this, we could generate JSON simply by doing `JSON.stringify`.
I know `snake_case` is a convention in Ruby, but here we're dealing with an external data format, and I think the additional complexity of snake case is difficult to justify.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 06:51:02
[cucumber/common] Comment on #1601
There is no such thing as JSON.stringify in ruby.
JSON.generate works with hashes only, not with objects. So we will have to write serializers anyway.
snake_case over camelCase was one of the point into generating DTOs over hashes.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 06:51:22
[cucumber/common] Comment on #1601
I think the message DTOs need to follow Ruby idioms. The whole point of adding the DTOs, for me, is to add a layer of indirection between the raw messages and client code, precisely so that we can do things like this.
Can you same more about the JSON.stringify use case? I didn't realise there were instances where we need to convert these objects back to raw JSON messages.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 06:51:23
[cucumber/common] mattwynne commented on aurelien-reeves's pull request
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 06:52:27
[cucumber/common] Comment on #935
<a href="https://github.com/vincent-psarga">@vincent-psarga</a> do you have any recollection of what's left to do on this?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 06:53:33
[cucumber/common] Comment on #1601
They need to be converted to JSON / NDJSON for the message formatter, and also for publishing to <a href="http://reports.cucumber.io">reports.cucumber.io</a>.
The only users of the Cucumber Messages Ruby API is us (the internal Cucumber-Ruby codebase). Going the roundabout way of using snake case seems completely unnecessary to me.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 06:54:19
[cucumber/common] Comment on #1601
> I think the message DTOs need to follow Ruby idioms. The whole point of adding the DTOs, for me, is to add a layer of indirection between the raw messages and client code, precisely so that we can do things like this.
>
> Can you same more about the JSON.stringify use case? I didn't realise there were instances where we need to convert these objects back to raw JSON messages.
The formatters output messages as ndjson
Also gherkin as a standalone provide a bin which allows to output ndjson from those messages
cf the CI here: <a href="https://github.com/cucumber/common/pull/1603">#1603</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 06:54:20
[cucumber/common] Comment on #331
Is anyone still interested in tackling this?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 06:57:22
[cucumber/common] Comment on #1601
> snake_case over camelCase was one of the point into generating DTOs over hashes.
That wasn't obvious to me. For me, the only benefit was that we can do `envelope.testStepFinished.timestamp` rather than having to do `envelope[:testStepFinished][:timestamp]`.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 06:58:36
[cucumber/common] Comment on #1601
Another benefit of DTOs is that we can throw an exception in the constructor if you pass an unsupported property.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 07:02:00
[cucumber/common] Comment on #1601
> The only users of the Cucumber Messages Ruby API is us (the internal Cucumber-Ruby codebase). Going the roundabout way of using snake case seems completely unnecessary to me.
There is at least one project which uses gherkin directly, and that may be impacted with changes in messages. Cf. <a href="https://github.com/cucumber/common/issues/1559">#1559</a>
Also, formatters may deal with messages directly. Thus, custom formatters may be impacted.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 07:03:34
[cucumber/common] Comment on #1601
> There is no such thing as JSON.stringify in ruby.
My bad. We could do this:
```
require 'json'
class Timestamp
attr_reader :seconds
attr_reader :nanos
def initialize(
seconds: 0,
nanos: 0
)
@seconds = seconds
@nanos = nanos
end
def to_hash
return {
seconds: seconds,
nanos: nanos
}
end
end
t = Timestamp.new(seconds: 10, nanos: 33)
puts t.to_hash.to_json
```
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 07:34:38
[cucumber/common] Comment on #935
> <a href="https://github.com/vincent-psarga">@vincent-psarga</a> do you have any recollection of what's left to do on this?
I'd say more or less everything :D
• port to other languages
• add tests based on messages to ensure every language behaves the same way
But the main issue here was that the progress formatter relies heavily on common code with other formatters, so as I mentioned earlier, having simply a "progress-formatter" library might not be such a good idea.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 09:30:55
[cucumber/common] Comment on #935
I think it would be fine to start by moving just cucumber-ruby's progress formatter into a separate (`cucumber-formatters-console`) gem in the common repo, and we could gradually migrate other formatters and languages in other PRs, once we've figured out a good way to make a shared test suite.
No need to try and boil the ocean; that's how things get stuck 😁
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 13:51:45
[cucumber/common] Comment on #1304
<a href="https://github.com/davidjgoss">@davidjgoss</a> hooks are filtered from the html report because jvm must emit them for each scenario (resulting in an drastic increase in file size).
So yeah, `getHook` should be nullable.
<a href="https://github.com/cucumber/cucumber-jvm/blob/main/core/src/main/java/io/cucumber/core/plugin/HtmlFormatter.java#L28">https://github.com/cucumber/cucumber-jvm/blob/main/core/src/main/java/io/cucumber/core/plugin/HtmlFormatter.java#L28</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 14:46:22
[cucumber/common] Comment on #1562
I THINK IT WORKS!!!?!?!?!
<a href="https://camo.githubusercontent.com/c968854c78261bab46c47ecc791aaa7900a5a54af09d1e9da21a89c5ef48edd0/68747470733a2f2f6d65646961302e67697068792e636f6d2f6d656469612f776b41316c316c45787a74776b2f67697068792d646f776e73697a65642d6d656469756d2e676966">https://camo.githubusercontent.com/c968854c78261bab46c47ecc791aaa7900a5a54af09d1e9da21a89c5ef48edd0/68747470733a2f2f6d65646961302e67697068792e636f6d2f6d656469612f776b41316c316c45787a74776b2f67697068792d646f776e73697a65642d6d656469756d2e676966</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 14:47:57
[cucumber/common] Comment on #1562
I had to manually `ln -s /usr/bin/chromium /usr/bin/chromium-browser` because puppeteer doesn't seem to be able to find `chromium` by itself. I don't know why that doesn't fail on other architectures too. I guess we need to add that to the docker environment image, or maybe there's a way to set it with an environment variable.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 14:48:57
[cucumber/common] Comment on #1562
<a href="https://github.com/ciaranmcnulty">@ciaranmcnulty</a> I'd appreciate you trying this out if you have a moment.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 14:59:02
[cucumber/common] Comment on #1562
I also think that `make clean && make` doesn't work reliably regarding installing `npm` packages - something's still not right there.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-09 15:40:16
[cucumber/common] Comment on #1562
<a href="https://user-images.githubusercontent.com/1000/121438506-057ae580-c97c-11eb-92a5-e42152a0c127.png">image</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 00:58:17
[cucumber/common] Comment on #935
> No need to try and boil the ocean; that's how things get stuck 😁
Indeed :)
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 01:31:11
[cucumber/common] Comment on #1555
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> would be usefull to have this for Java too (e.g. to format existing feature files), in general it would be nice to support a configuration file where one could configure some aspects e.g. indentation deep, indentation style (tab versus spaces), numer of new lines before/after a given scenario/data table/ ...
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 01:32:57
[cucumber/common] Comment on #1304
<a href="https://github.com/mpkorstanje">@mpkorstanje</a> ah thanks, that makes sense. Should be a straightforward fix then.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 02:04:28
[cucumber/common] Pull request opened by aurelien-reeves
<p>Add serialization capabilities to ruby DTOs</p>
<p>We must be able to serialize DTOs to json strings</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 02:14:49
[cucumber/common] Issue opened by aslakhellesoy
<p><strong>Summary</strong></p>
<p>Add a new <code>buildNumber</code> field on the <a href="https://github.com/cucumber/common/blob/main/messages/messages.md#ci">Ci</a> message.</p>
<p>This will allow tools like <a href="https://reports.cucumber.io/">Cucumber Reports</a> to display more details about a reports (We need it in Cucumber Reports' Report Collections)</p>
<p><strong>Expected Behavior</strong></p>
<p>The <code>Ci#buildNumber</code> field is set from a suitable CI tool-specific environment variable.</p>
<p><strong>Possible Solution</strong></p>
<p>• Add a new optional <code>buildNumber</code> field in the Cucumber Messages JSON schema for <code>Ci</code>
• Update the <a href="https://github.com/cucumber/common/tree/main/create-meta/testdata">test data in create-meta</a>.** Update <a href="https://github.com/cucumber/common/tree/main/create-meta">create-meta</a> to set this field for all CI servers that expose a build number.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 02:20:07
[cucumber/common] Comment on #1555
<a href="https://github.com/laeubi">@laeubi</a> this will be published as a command line tool that you can run as `npx @cucumber/gherkin-utils`. This requires Node.js to be installed.
I realise people who don't use Node.js might not have it installed, but it's relatively easy to install. I'm not going to prioritise porting this to Java, but if someone submits a PR for this we'd accept it. The PR would also have to extract the test data to reusable tests so we can ensure consistent behaviour between implementations. A bit like we've done in Gherkin and create-meta.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 02:35:55
[cucumber/common] Pull request ready for review by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 02:43:11
[cucumber/common] Comment on #1555
<a href="https://github.com/davidjgoss">@davidjgoss</a> <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> <a href="https://github.com/mattwynne">@mattwynne</a> could one of you review this please? Thanks!
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 03:02:28
[cucumber/common] Comment on #1555
Suggestion: rename it as `identStep` to remain consistent with the other function names
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 03:02:28
[cucumber/common] Comment on #1555
Suggested change
?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 03:02:28
[cucumber/common] Comment on #1555
Could we rename that single-letter variable?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 03:02:28
[cucumber/common] Comment on #1555
I was confused when I saw those two functions.
What do you think renaming `indentSpace` into `indent`, and `indent` into `indentHierarchy`?
Also the second function could be used in the first one for non-markdown syntax
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 03:02:29
[cucumber/common] aurelien-reeves commented on aslakhellesoy's pull request
I'll try to find some time to have proper tests
At the moment, here are a few comments after a first quick look at the code :)
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 04:52:16
[cucumber/common] Comment on #1555
What would you suggest renaming it to <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a>?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 04:54:39
[cucumber/common] Comment on #1555
Done - defaults to 'gherkin'
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 05:06:44
[cucumber/common] Comment on #1555
I've renamed `indent` to `keywordPrefix` and `indentSpace` to `spaces`
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 05:08:17
[cucumber/common] Comment on #1555
Renamed to `stepPrefix` to be consistent with `keywordPrefix`
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 05:31:33
[cucumber/common] Comment on #1555
`prettifiedFeature`?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 07:22:06
[cucumber/common] Comment on #1555
The current CLI interface isn't very shell friendly. Consider making it easy to do:
Output formatted std in to std out: `cat test.feature | gherkin-utils format`
Output formatted feature files to std out: `gherkin-utils format`
Format feature files in place: `gherkin-utils format --in-place [ FILE1, FILE2, FILE3.... ]`
Format selected feature files in place: `find <some thing> . -exec gherkin-utils format --in-place {} \;`
Convert to markdown in place: `gherkin-utils convert2md --in-place [ FILE1, FILE2, FILE3.... ]`
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 07:40:20
[cucumber/common] Comment on #1605
Deserialization of messages with arrays of other types of messages won't be easy without the addition of a complex layer which would analysis the DTOs and the schemas to find appropriate classes
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 08:56:35
[cucumber/common] Comment on #1555
I wonder if we could just call this CLI tool `gherkin` instead of `gherkin-utils`. We could still build/release it as a separate library to the parser code (`gherkin-cli` would follow a common JS idiom for the package name), but we can think of this as a user interface for the parser.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 09:05:11
[cucumber/common] Comment on #1555
The pretty formatter for TS/JS was <a href="https://github.com/jbpros/cucumber-pretty-formatter">recently given an official re-write</a> - could we not re-use that code here?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 09:05:11
[cucumber/common] mattwynne commented on aslakhellesoy's pull request
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 09:06:08
[cucumber/common] Comment on #1555
Or am I reading it wrong and this is for pretty-printing markdown?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 11:22:03
[cucumber/common] Comment on #1555
This library converts a <a href="https://github.com/cucumber/common/blob/main/messages/messages.md#gherkindocument">GherkinDocument</a> to a Gherkin or Markdown string, depending on what `Syntax` is passed as the 2nd argument.
The other pretty formatter has a different processing model - it prints lines as results are coming in. This library just prints the whole thing in one go.
I'd rather evolve this codebase to be able to print with results as well, but I'd rather handle that in a separate pull request.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-10 11:28:01
[cucumber/common] Comment on #1302
I'll get this up to date and then do a review - sorry it's gone unreviewed so long.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-11 04:12:13
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
mockito/mockito
<strong><a href="https://togithub.com/mockito/mockito/releases/v3.11.1"><code>v3.11.1</code></a></strong></p>
<p><a href="https://togithub.com/mockito/mockito/compare/v3.11.0...v3.11.1">Compare Source</a></p>
<p><em>Changelog generated by <a href="https://togithub.com/shipkit/shipkit-changelog">Shipkit Changelog Gradle Plugin</a></em></p>
<p><strong>3.11.1</strong></p>
<p>• 2021-06-11 - <a href="https://togithub.com/mockito/mockito/compare/v3.11.0...v3.11.1">3 commit(s)</a> by Charles Munger, dependabot[bot]
• Bump versions.bytebuddy from 1.11.1 to 1.11.2 <a href="https://togithub.com/mockito/mockito/pull/2322">(#2322)</a>
• Check package-privacy of method params <a href="https://togithub.com/mockito/mockito/pull/2318">(#2318)</a>
• Bump shipkit-auto-version from 1.1.16 to 1.1.17 <a href="https://togithub.com/mockito/mockito/pull/2317">(#2317)</a></p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
GitHub
2021-06-11 07:56:00
[cucumber/common] Pull request merged by mpkorstanje
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-11 10:16:22
[cucumber/common] davidjgoss commented on hWorblehat's pull request
This is looking good <a href="https://github.com/hWorblehat">@hWorblehat</a>, thanks for doing it and sorry it's been dormant for a while. I've made a couple of edits to bring it up to date with the changes on `main` since you raised.
Couple of observations:
• It's very customisable, down to what the query parameter names are and the function to redraw the URL. We _could_ consider dialling that down to make for a smaller albeit more opinionated change initially
• It would be neat if the URL bar of the browser also updated in real time - as a user my muscle memory wants to copy and paste from there. We could maybe look at doing that with <a href="https://developer.mozilla.org/en-US/docs/Web/API/History/replaceState">`history.replaceState`</a> but no rush to do that in this PR - would need to take care we don't cause issues with client-side routing in React apps etc.
Interested in <a href="https://github.com/mattwynne">@mattwynne</a> and <a href="https://github.com/aslakhellesoy">@aslakhellesoy</a>'s takes as well.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-11 10:25:15
[cucumber/common] Comment on #1605
Nah, here is a simple solution:
```
require 'json'
class Envelope
attr_reader :meta
def initialize(
meta: nil
)
@meta = meta
end
def self.from_hash(hash)
self.new(
meta: Meta.from_hash(hash[:meta])
)
end
end
class Meta
attr_reader :protocol_version
attr_reader :implementation
def initialize(
protocol_version: nil,
implementation: nil
)
@protocol_version = protocol_version
@implementation = implementation
end
def self.from_hash(hash)
self.new(
protocol_version: hash[:protocolVersion],
implementation: Product.from_hash(hash[:implementation])
)
end
end
class Product
attr_reader :name
attr_reader :version
def initialize(
name: nil,
version: nil
)
@name = name
@version = version
end
def self.from_hash(hash)
self.new(
name: hash[:name],
version: hash[:version]
)
end
end
describe '.from_hash' do
it 'reconstructs a deep object' do
# Normally we'd construct the hash with JSON.parse(json, {symbolize_names: true})
hash = {
meta: {
protocolVersion: '10.2.3',
implementation: {
name: 'cucumis',
version: '55.7'
}
}
}
envelope = Envelope.from_hash(hash)
expect(envelope.class).to eq Envelope
expect(envelope.meta.class).to eq Meta
expect(envelope.meta.implementation.class).to eq Product
puts envelope.inspect
end
end
```
We just need to generate a `.from_hash` class method on each class.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-11 10:44:37
[cucumber/common] Comment on #1605
And for arrays, something like this:
```
def self.from_hash(hash)
self.new(
object_array: (hash[:object_array] || []).map {|element| SomeClass.from_hash(element)},
primitive_array: (hash[:primitive_array] || [])
)
end
```
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-11 12:06:01
[cucumber/common] Comment on #1302
> It would be neat if the URL bar of the browser also updated in real time - as a user my muscle memory wants to copy and paste from there. We could maybe look at doing that with history.replaceState but no rush to do that in this PR - would need to take care we don't cause issues with client-side routing in React apps etc.
👍 to this. I really like it if you can copy/paste URLs for search results.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-11 12:07:19
[cucumber/common] Comment on #1302
> It's very customisable, down to what the query parameter names are and the function to redraw the URL. We could consider dialling that down to make for a smaller albeit more opinionated change initially
I've not read the code so I don't know how much complexity this adds, but I'm curious if there was a use-case in mind, like embedding this stuff in another app. In general making our stuff extensible / embed-able seems like a good ambition.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-11 13:20:14
[cucumber/common] Comment on #1605
I like the idea of adding (manually-written) specs for this behaviour in the DTO, even if the implementation of the solution is generated code.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-11 14:53:08
[cucumber/common] Pull request opened by mattwynne
<p><strong>Summary</strong></p>
<p>Ref <a href="https://github.com/cucumber/common/issues/1550">#1550</a> - attempt to switch to using VCS module paths within the monorepo instead of pointing to a subrepo.</p>
<p><strong>Details</strong></p>
<p>It doesn't work yet.</p>
<p><strong>Motivation and Context</strong></p>
<p>If we can use module paths that point to the directory within the cucumber/common repo, we can get rid of a whole load of complexity around creating subrepo mirrors.</p>
<p><strong>How Has This Been Tested?</strong></p>
<p><code>cd gherkin/go &amp;&amp; make</code></p>
<p><strong>Types of changes</strong></p>
<p>☐ Bug fix (non-breaking change which fixes an issue).
☐ New feature (non-breaking change which adds functionality).
☐ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-11 14:55:44
[cucumber/common] Pull request opened by mattwynne
<p><strong>Summary</strong></p>
<p>Change default Makefile for Go packages to rebuild if the <code>go.mod</code> file changes</p>
<p><strong>Motivation and Context</strong></p>
<p>I'm working on <a href="https://github.com/cucumber/common/issues/1550">#1550</a> and fiddling about with the <code>go.mod</code> files. It seems the Makefile needs this dependency expressed.</p>
<p><strong>Types of changes</strong></p>
<p>☑︎ Technical debt
☐ Bug fix (non-breaking change which fixes an issue).
☐ New feature (non-breaking change which adds functionality).
☐ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-11 14:59:43
[cucumber/common] Comment on #1608
i had a fiddle around but I can't seem to make this work. <a href="https://github.com/lonnblad">@lonnblad</a> <a href="https://github.com/jaysonesmith">@jaysonesmith</a> <a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> anyone care to take a look at it?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-11 15:00:34
[cucumber/common] Comment on #1608
The error I see is:
```
go: <a href="http://github.com/cucumber/common/messages/v16@v16.0.0">github.com/cucumber/common/messages/v16@v16.0.0</a>: missing <a href="http://github.com/cucumber/common/messages/go.mod">github.com/cucumber/common/messages/go.mod</a> and .../v16/go.mod at revision messages/v16.0.0
```
This combination seems to at least help it to figure out the right git tag. If I use `<a href="http://github.com/cucumber/common/messages/go/v16">github.com/cucumber/common/messages/go/v16</a>` as the string it seems to get even more confused:
```
go: <a href="http://github.com/cucumber/common/messages/go/v16@v16.0.0">github.com/cucumber/common/messages/go/v16@v16.0.0</a>: reading <a href="http://github.com/cucumber/common/messages/go/messages/go/go.mod">github.com/cucumber/common/messages/go/messages/go/go.mod</a> at revision messages/go/v16.0.0: unknown revision messages/go/v16.0.0
```
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-11 15:27:28
[cucumber/common] jaysonesmith approved mattwynne's pull request
I’m not super familiar with Make’s behaviors but it seems logical that if dependencies change that things should be rebuilt.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-11 15:34:02
[cucumber/common] aslakhellesoy approved mattwynne's pull request
LGTM!
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-11 15:40:04
[cucumber/common] Comment on #1605
Yeah we need specs for the deserialisation. I didn't modify the code generator to add those `.from_hash` class methods, the code I pasted above if just a proof of concept of what we _could_ generate.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-11 16:20:35
[cucumber/common] Pull request merged by mattwynne
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-11 17:57:48
[cucumber/common] Comment on #1608
There is no Git tag named `v16.0.0` in the monorepo. all tags are of the form `library/version`, so you need to use something like this:
```
<a href="http://github.com/cucumber/common/messages/v16">github.com/cucumber/common/messages/v16</a> messages/v16.0.0
```
I haven't tested if this is valid go module syntax. If it isn't, we have to start using a different naming scheme for tags - perhaps with a hyphen or dot instead of a slash,
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-12 02:43:09
[cucumber/common] Pull request opened by davidjgoss
<p><strong>Summary</strong></p>
<p>Prevent an error when rendering the failure of a hook where that hook has no <code>hook</code> message from which to find its location.</p>
<p><strong>Motivation and Context</strong></p>
<p>Fixes <a href="https://github.com/cucumber/common/issues/1304">#1304</a>.</p>
<p>cucumber-jvm filters out <code>hook</code> messages due to volume, so we need to handle that gracefully when rendering.</p>
<p><strong>How Has This Been Tested?</strong></p>
<p>• Unit test added
• Modified acceptance test data locally to test in storybook</p>
<p><strong>Types of changes</strong></p>
<p>☑︎ Bug fix (non-breaking change which fixes an issue).
☐ New feature (non-breaking change which adds functionality).
☐ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<p><strong>Checklist:</strong></p>
<p>☐ The change has been ported to Java.
☐ The change has been ported to Ruby.
☐ The change has been ported to JavaScript.
☐ The change has been ported to Go.
☐ The change has been ported to .NET.
☑︎ I've added tests for my code.
☐ My change requires a change to the documentation.
☐ I have updated the documentation accordingly.
☑︎ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-12 03:21:31
[cucumber/common] Pull request opened by davidjgoss
<p><strong>Summary</strong></p>
<p>Stop the anchor link for the step from overlapping the status icon.</p>
<p>(Must have missed this in the big styling PR last month.)</p>
<p><strong>Screenshots (if appropriate):</strong></p>
<p>Before:</p>
<p><a href="https://user-images.githubusercontent.com/3192745/121772865-20ae4680-cb70-11eb-9150-16a1ace5ebc9.png">image</a></p>
<p>After:</p>
<p><a href="https://user-images.githubusercontent.com/3192745/121772880-315ebc80-cb70-11eb-84d2-abdeed3bba2d.png">image</a></p>
<p><strong>Types of changes</strong></p>
<p>☑︎ Bug fix (non-breaking change which fixes an issue).
☐ New feature (non-breaking change which adds functionality).
☐ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<p><strong>Checklist:</strong></p>
<p>☐ The change has been ported to Java.
☐ The change has been ported to Ruby.
☐ The change has been ported to JavaScript.
☐ The change has been ported to Go.
☐ The change has been ported to .NET.
☐ I've added tests for my code.
☐ My change requires a change to the documentation.
☐ I have updated the documentation accordingly.
☑︎ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-12 05:44:52
[cucumber/common] Comment on #1302
Hi <a href="https://github.com/davidjgoss">@davidjgoss</a>, <a href="https://github.com/mattwynne">@mattwynne</a>,
Thanks for taking a look, and no worries about it being dormant - I've been distracted with other things anyway.
* * *
> It would be neat if the URL bar of the browser also updated in real time - as a user my muscle memory wants to copy and paste from there. We could maybe look at doing that with `history.replaceState` but no rush to do that in this PR - would need to take care we don't cause issues with client-side routing in React apps etc.
I actually started out with this approach, but using `history.pushState`. After playing with it I felt that the browser history got polluted too quickly, so switched to the current approach.
I'm not that familiar with browser APIs, and it looks like I missed `replaceState` as an option. I can add a change to do this next week - it should remove quite a bit of code.
* * *
> It's very customisable, down to what the query parameter names are and the function to redraw the URL. We could consider dialling that down to make for a smaller albeit more opinionated change initially
I didn't have a particular use case in mind, other than trying to keep cucumber-react as a component library that could easily be reused and embedded in other apps. Given that we're talking about a feature that messes with a property of the entire web page, it felt like being able to customize things like the names of the query parameters would be useful. If we decide to go with the `replaceState` option as above, that would remove a lot of the complexity in the code, and the need for customization, as we wouldn't need to separately render a URL outside of the window location.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-12 17:18:30
[cucumber/common] Comment on #1608
It seems to have ways of inferring the tag from the version string that I can't quite fathom yet. If you see the first error (`...at revision messages/v16.0.0`) it seems to have figured out the tag name and found it, but it has the wrong path to the actual `go.mod` file (`missing <http://github.com/cucumber/common/messages/go.mod`|github.com/cucumber/common/messages/go.mod`>)
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-12 17:19:08
[cucumber/common] Comment on #1608
In the second case it makes a different guess for the tag name, and can't find it in git: `unknown revision messages/go/v16.0.0`
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-13 23:10:09
[cucumber/common] Comment on #1605
Thanks <a href="https://github.com/aslakhellesoy">@aslakhellesoy</a>
I'll study that today 👍
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-14 00:57:13
[cucumber/common] Comment on #1608
go look for the module on the master branch.
But it does not exist (yet) because messages/go.mod still have `module <a href="http://github.com/cucumber/messages-go/v16">github.com/cucumber/messages-go/v16</a>`
with the following in `gherkin/go/go.mod` I was able to have a coherent error message:
```
<a href="http://github.com/cucumber/common/messages/go/v16">github.com/cucumber/common/messages/go/v16</a> e0d826edaa5d24e50910eaa689194a8b3340c907
# go: <a href="http://github.com/cucumber/common/messages/go/v16@v16.0.0-20210611214814-e0d826edaa5d">github.com/cucumber/common/messages/go/v16@v16.0.0-20210611214814-e0d826edaa5d</a>: parsing go.mod:
# module declares its path as: <a href="http://github.com/cucumber/common/messages/v16">github.com/cucumber/common/messages/v16</a>
# but was required as: <a href="http://github.com/cucumber/common/messages/go/v16">github.com/cucumber/common/messages/go/v16</a>
```
I think we will have to update the modules like the following (in `messages/go/go.mod`):
```
module <a href="http://github.com/cucumber/common/messages/go/v16">github.com/cucumber/common/messages/go/v16</a>
```
So we may need at least 2 PR to make it work:
The first one updating all go.mod replacing the module path from the subrepos to the appropriate monorepo sub-directory (for example `module <a href="http://github.com/cucumber/messages-go/v16">github.com/cucumber/messages-go/v16</a>` => `module <http://github.com/cucumber/common/messages/go/v16`|github.com/cucumber/common/messages/go/v16`>)
Then a second PR to update all the go `require` directive to point to the new location.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-14 01:09:31
[cucumber/common] Comment on #1610
Suggested change
Just in case?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-14 01:09:32
[cucumber/common] aurelien-reeves commented on davidjgoss's pull request
Could we take advantage of this PR to add a couple of unit tests here?
One test has been added to check the case of `Unknown location`. Could we check the two other `if`: the one when we find a `location`, and the one when we find a `javaMethod`?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-14 01:35:53
[cucumber/common] Comment on #1610
> Could we take advantage of this PR to add a couple of unit tests here?
Agreed, looking at that now.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-14 07:02:53
[cucumber/common] Comment on #1610
Implemented part of this, but with `sourceReference` according to the messages schema it must always be present on a `Hook`, so should be safe to omit the null check there?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-14 07:04:53
[cucumber/common] Comment on #1610
Added coverage of the other paths as suggested, and brought in <a href="https://testing-library.com/docs/react-testing-library/intro/">testing-library</a> to make the component testing a bit cleaner - long as we're happy with this I can do a follow-up to refactor a few other tests too.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-14 07:14:57
[cucumber/common] Comment on #1605
Here's a first draft of a generic solution.
The idea remain to generate as less code as possible when generating DTOs.
This is still draft. Some refactoring is needed.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-14 07:33:42
[cucumber/common] aurelien-reeves approved davidjgoss's pull request
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-14 07:42:47
[cucumber/common] Comment on #1610
Yes, maybe
I am always puzzled when trying to find the good balance between code safety vs being too much paranoid 😑
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-14 07:43:34
[cucumber/common] Pull request merged by davidjgoss
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-14 07:46:33
[cucumber/common] Comment on #1610
> Added coverage of the other paths as suggested,
Great, thanks!
> and brought in <a href="https://testing-library.com/docs/react-testing-library/intro/">testing-library</a> to make the component testing a bit cleaner
👍
> long as we're happy with this I can do a follow-up to refactor a few other tests too.
Once the CI is fixed, why not refactor a few tests
But I suggest not to refactor not too many of those, to avoid having a PR too big. WDYT?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-14 07:59:03
[cucumber/common] Comment on #1302
<a href="https://github.com/hWorblehat">@hWorblehat</a> fantastic, that sounds like a plan! Let me know if you need anything.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-15 01:26:42
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-15 06:10:51
[cucumber/common] Pull request opened by davidjgoss
<p>To get changes from <a href="https://github.com/cucumber/build/pull/34">cucumber/build#34</a></p>
<p>Should unblock <a href="https://github.com/cucumber/common/pull/1610">#1610</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-15 06:56:08
[cucumber/common] Comment on #1605
I've refactored the `Cucumber::Messages::Message` base class. It result in a bunch of small mixins.
I kept generated code as small as possible to avoid overloading the resulting DTOs.
I've added some rdoc to what seems to be a public API.
Unit tests could be added to the new mixins. But the new `message_spec.rb` already covers serialization and deserialization from a higher level.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-15 07:12:32
[cucumber/common] aurelien-reeves approved davidjgoss's pull request
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-15 07:19:17
[cucumber/common] Pull request merged by davidjgoss
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-15 07:19:59
[cucumber/common] Pull request closed by davidjgoss
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-15 10:52:16
[cucumber/common] Comment on #1550
Would be great if this eliminates the rsync'ing
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-15 12:09:11
[cucumber/common] Comment on #1302
Regarding `replaceState` / `pushState` I haven't played with this but I assume the pollution of the browser history is because we're searching as the user types, so each keypress results in a new URL in the browser?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-15 12:34:49
[cucumber/common] Comment on #1302
We don't update on every keystroke, but I think what we're looking to avoid with `replaceState` is the experience where you tweak a few filters or searches and then hit the back button expecting to land back on GitHub/Jenkins whatever but instead are just going back over each iteration of your filters.
(Correct me if I'm wrong <a href="https://github.com/hWorblehat">@hWorblehat</a>!)
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-15 14:36:24
[cucumber/common] Issue opened by mpkorstanje
<p>Currently everything depends on messages.</p>
<p>• This happens partially because messages contains the entire domain.
• And partially because we use messages as the Gherkin AST.</p>
<p>This is not great because:</p>
<p>• We have a serialization format at the core of our domain rather then in an adaptor.
• Because the Gherkin AST is the serialized format we can not utilize any patterns that make working with ASTs easier.
• When anything in the domain changes, everything has to be updated.
• Messages depends on Jackson. Due to it's nature as a de-serialization library <a href="https://search.maven.org/artifact/com.fasterxml.jackson.core/jackson-databind">Jackson gets frequent security updates</a>.</p>
<p>As a visual aid (note note quite accurate, I was sketching out something else):</p>
<p><a href="https://user-images.githubusercontent.com/6946919/122121915-731b8b80-ce2c-11eb-8ce3-ef5b0a2aa2c4.png">image</a></p>
<p>As such I would propose inverting the dependency and splitting messages into several smaller submodules. The modules are already hinted at by the comments in the proto file.</p>
<p><a href="https://github.com/cucumber/common/blob/f96daddbedd679969bee1609ac9a71e857752720/messages/messages.proto#L13-L43">common/messages/messages.proto</a></p>
<p>Lines 13 to 43 in </cucumber/common/commit/f96daddbedd679969bee1609ac9a71e857752720|f96dadd></p>
<p>By inverting the dependency:</p>
<p>• We push the serialization format out to the edges of the architecture. This allows for proper ports and adaptors design. Only root dependents (e.g. the Cucumber CLI) would have a dependency on messages.
• The impact of updating external dependency changes is limited to the root dependents (e.g. the Cucumber CLI).
• Changes that impact Gherkin won't impact execution related code.and vice-versa.</p>
<p><a href="https://user-images.githubusercontent.com/6946919/122126285-d22fcf00-ce31-11eb-8da5-754174e10038.png">image</a></p>
<p>Now I imagine the biggest perceived obstacle will be additional copying the domain objects into messages. However for example if the Gherkin AST was modelled as a tree of nodes using the visitor pattern it would be relatively straightforward to convert it to a message object.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-15 16:16:58
[cucumber/common] Comment on #1614
I agree with the thrust of these observations and the proposal. I want to think a bit more about whether I agree that we need to model this as multiple subdomains (parser, compiler, glue, execution, results) in separate modules or whether those stable types at the core of our domain (and, as you say, not depend on things like Jackson) could live in a single module.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-15 21:03:00
[cucumber/common] Comment on #1614
But isn't it the idea of Jackson and similar frameworks to have domain objects that could be (optionally) be serialized, just to prevent creating additional domain objects for serialization?
The consequence of your proposal would be that we have some kind of layer above messages? I'm just asking because Cucumber Plugins already has such a layer but it does not contains all information there and changes seem to be hard because of the discussion about what should be API or not, so in the end I ended up using the raw messages as they are offering the greatest flexibility.
This leads me to another point where I think if this is done and should be useful for different consumers (IDE, CLI, ...) the parsers and converter has all to be public accessible.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 00:40:51
[cucumber/common] Comment on #1608
`This comment was deleted`
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 00:44:47
[cucumber/common] Comment on #1614
<a href="https://github.com/mattwynne">@mattwynne</a> a single library would work too for the messages. As long as we don't treat the messages as our model.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 01:08:08
[cucumber/common] Pull request opened by davidjgoss
<https://github.com/cucumber/common/pull/1615|#1615 correct some bits of documentation for messages>
<p>• Remove mentions of protocol buffers from core language modules
• Correct first cucumber-js version to add support</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 01:42:34
[cucumber/common] Comment on #1605
traditionally this would be at the top as this is the first thing you read. Similar to what I have in SitePrism here: <a href="https://github.com/site-prism/site_prism/blob/main/lib/site_prism/dsl.rb#L11">https://github.com/site-prism/site_prism/blob/main/lib/site_prism/dsl.rb#L11</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 01:42:34
[cucumber/common] Comment on #1605
just having one private and then a line space is enough (Again aware this is auto-genned).
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 01:42:34
[cucumber/common] Comment on #1605
Stylistic question, flagging here.
Is there a reason we're not opting for the 1 class per file structure (I know this is auto-genned, but it still should be possible / desirable)
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 01:42:34
[cucumber/common] Comment on #1605
This can be heavily optimised to use `#dig`
`property.dig('items', $ref)` will return either `nil` or the value from the digging in. It's also slightly more optimal than doing double fetch iirc
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 01:42:34
[cucumber/common] Comment on #1605
not sure if we care about newline formatting? But flagging it here.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 01:42:35
[cucumber/common] luke-hill commented on aurelien-reeves's pull request
I couldn't review the codegen file proper. But flagged a few style suggestions.
I've not reviewed every bit as a lot looks repeated.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 03:42:06
[cucumber/common] Comment on #1605
The codegen script has been built like that, it is easier to remain consistent with other generated languages like typescript, and generating a single file for messages is really easier at the moment.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 03:53:58
[cucumber/common] Comment on #1605
Yes I know
Managing that in the ruby codegen erb template would add a lot of noise
I am not sure the benefit of having a proper automaticaly generated file would override that cost
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 03:55:57
[cucumber/common] Comment on #1605
Thanks <a href="https://github.com/luke-hill">@luke-hill</a> for the review!
The important it the PR is the serialization/deserialization capabilities.
I've just pushed refactored spec for those mixins.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 05:28:50
[cucumber/common] Comment on #1610
> Once the CI is fixed, why not refactor a few tests
Done - I think this is ready now.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 05:57:13
[cucumber/common] mpkorstanje approved davidjgoss's pull request
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 06:04:58
[cucumber/common] aurelien-reeves approved davidjgoss's pull request
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 06:07:36
[cucumber/common] Issue closed by davidjgoss
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 06:07:37
[cucumber/common] Pull request merged by davidjgoss
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 07:36:08
[cucumber/common] Pull request opened by mattwynne
<p>Ref <a href="https://github.com/cucumber/common/issues/1550">#1550</a></p>
<p>This PR changes each of the go module's paths to point to inside the monorepo instead of to the mirror subrepo.</p>
<p>This is to prepare the ground for <a href="https://github.com/cucumber/common/pull/1608">#1608</a> to update the references that point <em>to</em> these modules, once this PR has been merged to main, since Go default to look for the <code>go.mod</code> files on the repo's default branch.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 09:07:57
[cucumber/common] Comment on #1302
👍 Exactly
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 09:14:22
[cucumber/common] Comment on #1614
<a href="https://github.com/mpkorstanje">@mpkorstanje</a> I agree about not treating the messages as our model, they're just DTOs. What your post has given me pause for thought about is whether we should try to define a shared core domain model/models for Cucumber though. Like that maybe messages is just filling a gap right now, that should actually be filled by a proper domain model.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 09:15:43
[cucumber/common] Pull request merged by mattwynne
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 09:37:36
[cucumber/common] aurelien-reeves approved mattwynne's pull request
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 09:37:36
[cucumber/common] Comment on #1608
Wooot!!
Don't forget to update CHANGELOG before merging this, with an entry under `Changed` so we remember to make a major release next time.
And then I suppose we can archive all the mirrors and remove the mirroring stuff from the build? Including the executable we run to do the actual mirroring.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 09:40:27
[cucumber/common] Comment on #1608
> Wooot!!
>
> Don't forget to update CHANGELOG before merging this, with an entry under `Changed` so we remember to make a major release next time.
>
> And then I suppose we can archive all the mirrors and remove the mirroring stuff from the build? Including the executable we run to do the actual mirroring.
Changelog have already been updated as part of <a href="https://github.com/cucumber/common/pull/1616">#1616</a>
Maybe just add references to the PR into those?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 09:59:21
[cucumber/common] Comment on #1614
Ah. Yes you are definitely right that the messages are filling in a gap of domain objects. However we should not introduce a monolithic domain. Rather each module has it's own public API and only uses another modules public API. The domain objects can be part of this public API. For example:
<a href="https://user-images.githubusercontent.com/6946919/122258977-f17e3900-ced1-11eb-91a0-88c6da339a4a.png">image</a>
In this case we have two modules. A `compiler` and `parser` module. The `Parser`, `Compiler`, `AST` and `Pickle` classes are all part of it's public API. The AST and Pickle are also domain objects.
Internally each module can use patterns such as "Ports and Adapters" but I don't think this pattern is efficient for small libraries that do not have any input/output. Rather this is something you'd apply to utilities that use these libraries such as the Gherkin CLI because these have to deal with input and output where as the `parser` and `compiler` modules do not.
For example:
<a href="https://user-images.githubusercontent.com/6946919/122261234-7702e880-ced4-11eb-81df-a2f139fefd97.png">image</a>
Now this brings us to the next problem. The current implementation has to deal with input and output just to pass the shared test suite. This was a pragmatic choice, but it also means that as a module gherkin drags in more then needed.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 11:04:27
[cucumber/common] Pull request ready for review by mattwynne
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 11:05:01
[cucumber/common] Pull request merged by mattwynne
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 11:06:19
[cucumber/common] Comment on #1608
I think having the CHANGELOGs reference the main issue <a href="https://github.com/cucumber/common/issues/1550">#1550</a> is good enough. Do you agree?
We'll need to do the new major releases of messages and gherkin as part of <a href="https://github.com/cucumber/godog/issues/389">cucumber/godog#389</a> I guess.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 11:12:26
[cucumber/common] Comment on #1608
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> I've updated the checklist at the top of <a href="https://github.com/cucumber/common/issues/1550">#1550</a> - can you take a look and see if you agree with the remaining steps?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 11:15:38
[cucumber/common] Comment on #1544
<a href="https://github.com/gitissuepost">@gitissuepost</a> can you put the sample project on GitHub please, in a public repo? That will make it easier for people to read it.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 11:16:32
[cucumber/common] Comment on #1544
The exception is:
```
Exception in thread "main" io.cucumber.gherkin.ParserException$CompositeParserException: Parser errors:
(19:1): expected: #EOF, #Language, #TagLine, #FeatureLine, #Comment, #Empty, got 'Feature: Visit career guide page in <a href="http://career.guru99.com">career.guru99.com</a>'
(21:3): expected: #EOF, #Language, #TagLine, #FeatureLine, #Comment, #Empty, got 'Scenario: Visit <a href="http://career.guru99.com">career.guru99.com</a>'
(22:5): expected: #EOF, #Language, #TagLine, #FeatureLine, #Comment, #Empty, got 'Given I am on <a href="http://career.guru99.com">career.guru99.com</a>'
(23:7): expected: #EOF, #Language, #TagLine, #FeatureLine, #Comment, #Empty, got '|a|b|c|'
(24:7): expected: #EOF, #Language, #TagLine, #FeatureLine, #Comment, #Empty, got '|a1|b1|c1|'
(25:5): expected: #EOF, #Language, #TagLine, #FeatureLine, #Comment, #Empty, got 'When I click on career guide menu'
(26:7): expected: #EOF, #Language, #TagLine, #FeatureLine, #Comment, #Empty, got '|a|b|c|'
(27:7): expected: #EOF, #Language, #TagLine, #FeatureLine, #Comment, #Empty, got '|a1|b2|c1|'
(28:5): expected: #EOF, #Language, #TagLine, #FeatureLine, #Comment, #Empty, got 'Then I should see career guide page'
(29:7): expected: #EOF, #Language, #TagLine, #FeatureLine, #Comment, #Empty, got '|a|b|c|'
(30:7): expected: #EOF, #Language, #TagLine, #FeatureLine, #Comment, #Empty, got '|a1|b1|c1|'
at io.cucumber.gherkin.Parser.addError(Parser.java:159)
at io.cucumber.gherkin.Parser.matchTokenAt_0(Parser.java:549)
at io.cucumber.gherkin.Parser.matchToken(Parser.java:344)
at io.cucumber.gherkin.Parser.parse(Parser.java:138)
at io.cucumber.gherkin.Parser.parse(Parser.java:119)
at io.cucumber.gherkin.Parser.parse(Parser.java:115)
at parser.ParserTest.main(ParserTest.java:23)
```
This looks to be trying to parse some Gherkin that's different to what's shown in the comment above.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 11:18:46
[cucumber/common] Comment on #1544
Pulled from the sample project zip file, this is the feature that fails to parse:
```
#Author: <a href="mailto:your.email@your.domain.com">your.email@your.domain.com</a>
#Keywords Summary :
#Feature: List of scenarios.
#Scenario: Business rule through list of steps with arguments.
#Given: Some precondition step
#When: Some key actions
#Then: To observe outcomes or validation
#And,But: To enumerate more Given,When,Then steps
#Scenario Outline: List of steps for data-driven as an Examples and <placeholder>
#Examples: Container for s table
#Background: List of steps run before each of the scenarios
#""" (Doc Strings)
#| (Data Tables)
#@ (Tags/Labels):To group Scenarios
#<> (placeholder)
#""
## (Comments)
#Sample Feature Definition Template
Feature: Visit career guide page in <a href="http://career.guru99.com">career.guru99.com</a>
Scenario: Visit <a href="http://career.guru99.com">career.guru99.com</a>
Given I am on <a href="http://career.guru99.com">career.guru99.com</a>
|a|b|c|
|a1|b1|c1|
When I click on career guide menu
|a|b|c|
|a1|b2|c1|
Then I should see career guide page
|a|b|c|
|a1|b1|c1|
```
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 11:19:54
[cucumber/common] Comment on #1544
I'm confused about why that doesn't parse, it looks OK to me.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 12:21:34
[cucumber/common] Comment on #1544
The `TokenMatcher` is language sensitive. At glance I'd say that it hasn't been configured for the right language.
```
TokenMatcher matcher = new TokenMatcher("no");
```
That should probably be `"en"`.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 13:13:43
[cucumber/common] Comment on #1605
This surprising behaviour.
I would expect:
• JSON primitive => Ruby primitive
• JSON object => Ruby object
• JSON array => Ruby array
Here however, you're turning a JSON object into a Ruby array.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 13:13:43
[cucumber/common] Comment on #1605
This won't work. You cannot assume that the property name has a corresponding message type. For example, one of the messages has a property called `implementation` (of type `Product`), but here you're trying to access a non-existent constant called `Implementation`
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 13:13:44
[cucumber/common] aslakhellesoy requested changes on aurelien-reeves's pull request
You've gone for an implementation where the deserialisation logic has no knowledge of the schema and tried to implement logic in the `Deserialization` module that will figure it out with reflection.
This isn't going to work.
The deserialization logic has to know about the schema - either by loading the schema at runtime, or by generating deserialization code that knows about the schema.
I wouldn't recommend the runtime approach, which is why I suggested the code generation approach in <a href="https://github.com/cucumber/common/pull/1605#issuecomment-859731059">#1605 (comment)</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 14:27:33
[cucumber/common] Comment on #1605
This implementation will return a hash with `nil` values for fields that are `nil`.
This in turn will include all fields in the JSON, which we want to avoid because it increases message size considerably.
Unfortunately `JSON.generate` does not provide any options to not output `nil` values as `null`. We therefore need a `to_hash` implementation that only includes non-nil values.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 15:25:32
[cucumber/common] Pull request opened by aslakhellesoy
<p>This PR addresses the feedback I gave in my review of <a href="https://github.com/cucumber/common/pull/1605">#1605</a></p>
<p>All serialization/deserialization logic is implemented in generated classes. The message superclass with reflection logic is gone - I don't think it would have been possible to implement this properly with reflection unless we load the JSON schema at runtime to get the types for various fields (we don't want that - it would be complicated).</p>
<p>I've also updated the <code>NdjsonToMessageEnumerator</code> to emit proper <code>Envelope</code> objects and updated the Ruby demo formatter accordingly.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 15:50:58
[cucumber/common] Comment on #1605
I've created <a href="https://github.com/cucumber/common/pull/1617">#1617</a> which addresses the feedback I gave in this PR.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 16:25:02
[cucumber/common] Comment on #1544
Yes, that feature file does not look Norwegian :flag_no: at all to me :-)
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-16 16:31:07
[cucumber/common] Comment on #1550
I disagree with `use symlinks instead of rsync and delete rsync scripts/functions` - symlinks work poorly in git.
We may be able to replace some of the copied files with _relative file paths_ within the repo, but we might not be able to do it for all rsync'ed files.
For example, I think `.mocharc.json` files must be in the same directory as `package.json` - there might be more cases like that.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 00:23:50
[cucumber/common] Comment on #1550
> symlinks work poorly in git
Is that a Windows issue?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 02:27:32
[cucumber/common] Comment on #1617
When I added support for serialization/deserialization of Array fields, I discovered a bug in the schema for `TestStep`, so I fixed that too (all Array fields should be `required` so that they default to an empty array).
This had some ripple-effect to other parts of the monorepo (compatibility-kit, fake-cucumber, react, json-to-messages)
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 02:41:56
[cucumber/common] Comment on #1550
I thought it was a Windows issue (used to be at least), but have no simple way of verifying if it works now. Maybe it does now: <a href="https://github.community/t/windows-symlinks-do-they-play-well-with-git-projects/730">https://github.community/t/windows-symlinks-do-they-play-well-with-git-projects/730</a>
If symlinks work on Windows too, it would certainly simplify things!
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 02:43:21
[cucumber/common] Pull request merged by aslakhellesoy
<https://github.com/cucumber/common/pull/1615|#1615 correct some bits of documentation for messages>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 03:01:38
[cucumber/common] Comment on #1550
I have similar vague memories :)
I think under WSL2 it's fine but no idea how it works on 'real' Windows nowadays
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 03:02:36
[cucumber/common] Comment on #1550
If we don't have separate repos in future though I agree that we should minimise the use of symlinks in favour of relative paths
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 04:04:29
[cucumber/common] Comment on #1544
Maybe we could improve the API a bit here. Something like:
```
Parser parser = Parser.builder()
// Either
.detectLanguageFromFeature()
// Or
.defaultLanguage("en")
.build()
```
Currently the `detectLanguageFromFeature` is done by Cucumber JVM by looking for the `# language:` pragma. This has been around for so long now that perhaps we should make it part of the specification.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 04:08:12
[cucumber/common] Comment on #1544
Or maybe we can some up with something that _just works_TM.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 05:45:08
[cucumber/common] Comment on #1302
OK <a href="https://github.com/davidjgoss">@davidjgoss</a>, <a href="https://github.com/mattwynne">@mattwynne</a>,
This is ready for you to take a look at again now (assuming CI passes). Searching or updating the status filters replaces the URL in the bowser bar, but doesn't add an additional history element each time.
There's still a fair amount of customizability, but it's all encapsulated in the `SearchQueryContext` module and everything has sensible (IMO) defaults:
• Continuing to use `QueriesWrapper` as before, with either a string `query` prop or no such prop, will result in the old behaviour (i.e. no interaction with document URL). Hence, any third party uses of cucumber/react in the wild should continue to work without modification.
• The `main.tsx` in `html-formatter` passes the result of `searchFromURLParams()` into `QueriesWrapper` props, which binds the search and filters to browser URL query parameters, using default parameter names.
• Advanced uses (e.g. custom URL parameter names, binding the search to some other object in an embedding app) are catered for by passing options to `searchFromURLParams(...)` or instead manually setting the `{query, hiddenStatuses, onSearchQueryUpdated}` props on `QueriesWrapper`.
If you still think there's too much flexibility here, I'm happy to defer to your judgement and take it out. However, I'd always make that case that a component library shouldn't be too opinionated about how it's being used, particularly WRT interaction with global state such as the browser URL.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 08:46:42
[cucumber/common] Pull request opened by aslakhellesoy
<p>This is a fix for <a href="https://github.com/cucumber/common/issues/1173">#1173</a>.</p>
<p>Unexpanded:</p>
<p><a href="https://user-images.githubusercontent.com/1000/122430436-82fdb180-cf8b-11eb-8958-0f58f023b2a4.png">Screenshot 2021-06-17 at 16 41 07</a></p>
<p>Expanded:</p>
<p><a href="https://user-images.githubusercontent.com/1000/122430460-88f39280-cf8b-11eb-83d1-119b3fd87bd8.png">Screenshot 2021-06-17 at 16 41 37</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 09:17:54
[cucumber/common] Comment on #1544
Thank you all for your feedback. I didn't check that the TokenMatcher was initialized for Norwegian language. By setting it to "en", I am able to execute the script perfectly. However, i couldn't find any suitable method to read the step of a specific scenario.
<a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> : For time being I am closing this ticket. I will explore more to see if it's breaking some where and we can reopen this /log a new issue at that time.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 09:17:54
[cucumber/common] Issue closed by gitissuepost
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 09:25:56
[cucumber/common] davidjgoss commented on aslakhellesoy's pull request
Looks good!
One query around image widths. At the moment our only constraint is:
```
max_width: 100%
```
So if you have a big image (as in this example, but also common with browser screenshots) it will just go as wide as your screen and dominate the page.
Should we have a reasonable upper width limit e.g. 1200px? If so, should that be different within the examples table?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 09:41:39
[cucumber/common] Comment on #1619
Is it a problem that the image takes the full width? It's collapsed by default.
BTW, my change in the CCK seems to have triggered a test failure in `json-formatter` - I need to investigate that.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 09:43:27
[cucumber/common] Comment on #1619
> Is it a problem that the image takes the full width? It's collapsed by default.
Yeah I think that's a fair point, you're opting into it. Sold!
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 09:43:36
[cucumber/common] davidjgoss approved aslakhellesoy's pull request
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 09:47:29
[cucumber/common] Comment on #1617
One thing I liked about Aurellien's iteration was that he'd included RDoc with examples of calls to the API. I know we have RSpec examples but I'd like to see some docs on these methods too.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 09:48:57
[cucumber/common] Comment on #1617
I love that we're keeping the underscore / camelCase stuff out of the runtime code!
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 09:49:54
[cucumber/common] Comment on #1617
Yeah put them in. if you want to. I was focussed on making the tests pass :-)
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 09:50:10
[cucumber/common] Comment on #1617
I can't easily comprehend what each leg of the logic is doing here. Can we extract some methods?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 09:53:41
[cucumber/common] Comment on #1617
The first one is for arrays of objects
The second one is for objects
The third is for primitives and arrays of primitives.
I agree we should extract a method up into the `Ruby` class
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 13:15:29
[cucumber/common] Comment on #1423
Sorry for late reply <a href="https://github.com/fslev">@fslev</a>, you should see this fixed in the current version of cucumber-jvm - text attachments are now properly contained and scrollable horizontally.
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 13:15:31
[cucumber/common] Issue closed by davidjgoss
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 16:07:11
[cucumber/common] Issue closed by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 16:07:12
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-17 16:37:13
[cucumber/common] Comment on #1550
<a href="https://github.com/gasparnagy">@gasparnagy</a> <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> <a href="https://github.com/enkessler">@enkessler</a> can you add anything about the symlinks situation on Windows these days?
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-18 00:21:43
[cucumber/common] Comment on #1550
<a href="https://github.com/mattwynne">@mattwynne</a> I did a quick check. you can create symlinks, but only with admin privileges. so with normal console where you would do a git pull it does not work.
How does git support symlinks? Could anyone point me to a git repo that had symlinks inside - i would try to see what happens if i pull...
btw the win command for creating symlinks is
```
mklink symlink-file target-file
```
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-18 00:27:18
[cucumber/common] Comment on #1550
It seems you can enable creating symlinks without admin, but it requires you to enable developer mode (i don't have that enabled... and never heard about it yet). Unfortunately developer mode enables a bunch of other things, like SSH server, that you might not want to have on a laptop for example...
<a href="https://docs.microsoft.com/en-us/windows/apps/get-started/enable-your-device-for-development">https://docs.microsoft.com/en-us/windows/apps/get-started/enable-your-device-for-development</a>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-18 09:33:33
[cucumber/common] Comment on #1550
<p><a href="https://github.com/gasparnagy">@gasparnagy</a> try <a href="https://github.com/mattwynne/symlink-test">https://github.com/mattwynne/symlink-test</a> ?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-18 09:52:06
[cucumber/common] Comment on #1550
<p>You should be able to do:</p>
<pre><code>&gt; type link-to-readme.md
# symlink-test
</code></pre>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-20 07:24:27
[cucumber/common] Issue opened by iamgovindthakur
<p><strong>Summary</strong></p>
<p>Hi I have a Data like in the form of List<List< String >><br />
[[fName, lName, email, sex, mob, address], [Easter, Bristowe, <a href="mailto:ebristowe0@dropbox.com">ebristowe0@dropbox.com</a>, F, <tel:551-356-9751|551-356-9751>, 2894 Sutherland Park], [Raquela, MacGragh, <a href="mailto:rmacgragh1@ocn.ne.jp">rmacgragh1@ocn.ne.jp</a>, F, <tel:379-224-2320|379-224-2320>, 2599 Tennyson Trail], [Malinde, Minshaw, <a href="mailto:mminshaw2@time.com">mminshaw2@time.com</a>, F, <tel:106-223-2325|106-223-2325>, 1 Old Shore Street], [Rebekah, Tommen, <a href="mailto:rtommen3@dot.gov">rtommen3@dot.gov</a>, F, <tel:780-243-7802|780-243-7802>, 145 Briar Crest Street], [Norby, Devil, <a href="mailto:ndevil4@booking.com">ndevil4@booking.com</a>, M, <tel:101-322-0317|101-322-0317>, 5849 Vermont Way], [Halimeda, Baskwell, <a href="mailto:hbaskwell5@yellowpages.com">hbaskwell5@yellowpages.com</a>, F, <tel:615-791-6817|615-791-6817>, 13362 Forest Place], [Claudia, Ormshaw, <a href="mailto:cormshaw6@wunderground.com">cormshaw6@wunderground.com</a>, F, <tel:761-913-3213|761-913-3213>, 17711 Randy Trail], [Hortensia, McClenaghan, <a href="mailto:hmcclenaghan7@usnews.com">hmcclenaghan7@usnews.com</a>, F, <tel:141-855-3342|141-855-3342>, 3689 Old Gate Center], [Jerome, Galletly, <a href="mailto:jgalletly8@chicagotribune.com">jgalletly8@chicagotribune.com</a>, M, <tel:801-951-1393|801-951-1393>, 02 Monterey Lane], [Miller, Aiston, <a href="mailto:maiston9@histats.com">maiston9@histats.com</a>, M, <tel:641-793-5124|641-793-5124>, 814 Straubel Lane]]</p>
<p>I want the above data to get converted into DataTable with the help of TableConverter by using code like this</p>
<pre><code>public static DataTable.create(List&lt;List&lt;String&gt;&gt; raw, TableConverter tableConverter) {
return new DataTable(copy(requireRectangularTable(raw)), tableConverter);
}
</code></pre>
<p>Later I want to fetch the data from generated DataTable like this</p>
<pre><code>for (Map&lt;String, String&gt; users : data.asMaps(String.class, String.class)) {
System.out.println(users.get("email"));
}
</code></pre>
<p>Please help me implementing the above logic. As I don't know what to give in second argument of DataTable.create() function.<br />
Any help will be appreciated. I searched everywhere I did not get help from a single one. And please provide the documentation of all your functions with examples so that it will be self understanding.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-20 08:14:44
[cucumber/common] Pull request opened by mpkorstanje
<p>The current issue templates are mostly ignored. The ones provided by github<br />
don't tend to be so much.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-20 08:20:26
[cucumber/common] Comment on #1620
<p>I can't really work out what you're trying to do. If you are using cucumber-jvm you shouldn't have to use any of this. You can define the list in the the step definition and Cucumber will convert the table.</p>
<pre><code> @Given("a datatable as a list of maps:")
public void a_data_table(List&lt;Map&lt;String, String&gt;&gt; table){
}
</code></pre>
<p>Have a look at:</p>
<p><a href="https://github.com/cucumber/cucumber-jvm/tree/main/java#data-tables">https://github.com/cucumber/cucumber-jvm/tree/main/java#data-tables</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-20 08:20:26
[cucumber/common] Issue closed by mpkorstanje
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-21 01:56:19
[cucumber/common] Comment on #1621
<p>Can we mention something here about having either a repo or an MVCE?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-21 02:14:48
[cucumber/common] Comment on #1621
<p>Unfortunately most Cucumber users have a limited amount of experience with either. I'm afraid that by asking for anything specific will just get us nothing at all instead.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-21 08:59:46
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
mockito/mockito
<strong><a href="https://togithub.com/mockito/mockito/releases/v3.11.2"><code>v3.11.2</code></a></strong></p>
<p><a href="https://togithub.com/mockito/mockito/compare/v3.11.1...v3.11.2">Compare Source</a></p>
<p><em>Changelog generated by <a href="https://togithub.com/shipkit/shipkit-changelog">Shipkit Changelog Gradle Plugin</a></em></p>
<p><strong>3.11.2</strong></p>
<p>• 2021-06-21 - <a href="https://togithub.com/mockito/mockito/compare/v3.11.1...v3.11.2">5 commit(s)</a> by dependabot[bot]
• Bump versions.bytebuddy from 1.11.2 to 1.11.3 <a href="https://togithub.com/mockito/mockito/pull/2333">(#2333)</a>
• Bump assertj-core from 3.20.0 to 3.20.1 <a href="https://togithub.com/mockito/mockito/pull/2332">(#2332)</a>
• Bump org.eclipse.osgi from 3.16.200 to 3.16.300 <a href="https://togithub.com/mockito/mockito/pull/2330">(#2330)</a>
• Bump assertj-core from 3.19.0 to 3.20.0 <a href="https://togithub.com/mockito/mockito/pull/2329">(#2329)</a>
• Bump shipkit-auto-version from 1.1.17 to 1.1.19 <a href="https://togithub.com/mockito/mockito/pull/2328">(#2328)</a></p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-21 10:00:13
[cucumber/common] Comment on #1441
<p>Some work here <a href="https://github.com/cucumber/common/pull/1621">#1621</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-21 10:10:17
[cucumber/common] Pull request merged by mpkorstanje
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-21 10:19:26
[cucumber/common] Issue closed by mattwynne
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-21 10:19:36
[cucumber/common] Comment on #1441
<p>🎉 thanks Rien!</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-21 10:26:42
[cucumber/common] Comment on #1441
<p>??</p>
<p>I didn't fix the PR template. Just the issue ones.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-21 11:04:14
[cucumber/common] Comment on #1441
<p>Ho ho whoops I didn't read your PR carefully enough!</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-21 11:04:15
[cucumber/common] Issue reopened by mattwynne
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-21 11:40:20
[cucumber/common] Pull request merged by mpkorstanje
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-21 13:33:19
[cucumber/common] Issue opened by sashokbg
<p>Hello, I am spending some of my free time working on the cukedoctor project (<a href="https://github.com/rmpestano/cukedoctor">https://github.com/rmpestano/cukedoctor</a>) which aims to add AsciiDoc format features to cucumber.</p>
<p>The main idea is to parse the the JSON output of the cucumber tests and compile them to adoc pages.</p>
<p>The project still uses the old JSON format and I am trying to migrate it to the new cucumber messaging system.<br />
I am however unable to find examples on how to properly parse the messages and would appreciate any guides on how to do it.</p>
<p>I have seen the json formatter (written in go) but I am having a very hard team reading go.<br />
I have also seen the json schema for the messages but no examples written in Java on how to use them.</p>
<p>Best Regards</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-21 14:10:49
[cucumber/common] Pull request opened by artysidorenko
<p><strong>Summary</strong></p>
<p>Updates DataTable with a new print method allowing to configure custom indention, and skip escaping the cell contents.</p>
<p><strong>Details</strong></p>
<p>• makes the <code>TablePrinter</code> class configurable, via a new <code>TablePrinterBuilder</code> class.
• adds a new method <code>printFormatted()</code> to the <code>DataTable</code> class, as it already includes similar publicly-exposed <code>print()</code> methods
• above method configures the <code>TablePrinter</code> with custom indent + removing cell-escaping.</p>
<p>(I'm still not too familiar with the codebase; the above felt like reasonable decisions, but am happy to incorporate feedback if you prefer doing it different)</p>
<p><strong>Motivation and Context</strong></p>
<p>Relates to <a href="https://github.com/cucumber/cucumber-jvm/issues/2320">issue-2320 on cucumber-jvm</a>.</p>
<p>Ultimate aim is to include data-tables in the printed test output. E.g. in the following feature, current behaviour prints out only the step definition, but it skips the info in the table.</p>
<pre><code>Given I have a list of people
| NAME | AGE |
| John | 25 |
| Sally | 28 |
</code></pre>
<p><strong>How Has This Been Tested?</strong></p>
<p>Added 2 unit tests. <code>can_print_table_with_escape_characters_regular</code> confirms the escape-cell behaviour of the original <code>print()</code> method. And <code>can_print_table_with_escape_characters_formatted</code> tests the behaviour of the new <code>printFormatted()</code> method.</p>
<p><strong>Types of changes</strong></p>
<p>☐ Bug fix (non-breaking change which fixes an issue).
☑︎ New feature (non-breaking change which adds functionality).
☐ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<p><strong>Checklist:</strong></p>
<p>☑︎ The change has been ported to Java.
☐ The change has been ported to Ruby.
☐ The change has been ported to JavaScript.
☐ The change has been ported to Go.
☐ The change has been ported to .NET.
☑︎ I've added tests for my code.
☐ My change requires a change to the documentation.
☐ I have updated the documentation accordingly.
☑︎ I have updated the CHANGELOG accordingly.</p>
<p>(don't think it requires documentation/porting to other languages as it's a small change?)</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-21 20:33:44
[cucumber/common] Pull request opened by willmac321
<p>Add an en-tx (US Texan) localisation :) Thank's y'all!</p>
<p><strong>Summary</strong></p>
<p>Add a US Texan localization to gherkin-languages.json</p>
<p><strong>Motivation and Context</strong></p>
<p>Regarding the question about solving, it makes it so I can understand y'all better! Also for the memes.</p>
<p><strong>How Has This Been Tested?</strong>
<strong>Screenshots (if appropriate):</strong>
<strong>Types of changes</strong></p>
<p>☐ Bug fix (non-breaking change which fixes an issue).
☐ New feature (non-breaking change which adds functionality).
☐ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<p><strong>Checklist:</strong></p>
<p>☐ The change has been ported to Java.
☐ The change has been ported to Ruby.
☐ The change has been ported to JavaScript.
☐ The change has been ported to Go.
☐ The change has been ported to .NET.
☐ I've added tests for my code.
☐ My change requires a change to the documentation.
☐ I have updated the documentation accordingly.
☐ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-21 23:47:17
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-22 08:51:47
[cucumber/common] Comment on #1627
<p><a href="https://github.com/ciaranmcnulty">@ciaranmcnulty</a> and I paired on this today for half and hour and made a little start. We're going to try and work on <code>/messages/javascript</code> first as a proof of concept.</p>
<p>Here's what we discovered so far:</p>
<p>• ✨ We can use <code>docker build --file ../../Dockerfile</code> to use a single shared <code>Dockerfile</code> in the monorepo root, and still have the build context be local to the package/language folder.
• 🤔 Should we copy files or map volumes? We don't have any long-running processes (like a web server) where you'd want to change files in your editor and see the changes, but we do need to allow for people doing TDD on these packages and make sure they run specific tests and get fast feedback. We want to try both and see what the performance and experience is like.
• 🤔 The codegen target for <code>src/messages.ts</code> needs Ruby, so maybe the codegen step needs to be done outside of the language build, in another container.
• 🤔 That same codegen target expects to be run in the context of the monorepo (it looks for <code>../jsonschema/scripts/codegen.rb</code>) which it can't see if we only copy files from the local folder. Maybe it makes more sense to map a volume of the entire monorepo?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-22 09:12:39
[cucumber/common] Comment on #1627
<p>> Should we copy files or map volumes?</p>
<p>We'll have to map at least some volumes, so that changes made by the build process are reflected on the host (to be committed back to the repo). My first attempt would be to copy files into the containers (for speed on non-Linux, and to do optimisations like only running installers sometimes). The proof will be in the real-world performance</p>
<p>> The codegen target for src/messages.ts needs Ruby, so maybe the codegen step needs to be done outside of the language build, in another container.</p>
<p>Other tasks like rsync etc. can be handled by a top-level container that spawns the other containers (which can be allowed by mounting docker.sock into the container)</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-23 00:36:38
[cucumber/common] Pull request opened by renovate[bot]
<https://github.com/cucumber/common/pull/1628|#1628 Update dependency @testing-library/react to v12>
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
testing-library/react-testing-library
<strong><a href="https://togithub.com/testing-library/react-testing-library/releases/v12.0.0"><code>v12.0.0</code></a></strong></p>
<p><a href="https://togithub.com/testing-library/react-testing-library/compare/v11.2.7...v12.0.0">Compare Source</a></p>
<p><strong>Bug Fixes</strong></p>
<p>• Bump testing-library/dom to v8 alpha (<a href="https://togithub.com/testing-library/react-testing-library/issues/923">#923</a>) (<a href="https://togithub.com/testing-library/react-testing-library/commit/770246e5cf15593bee96de5ce8b43305826c0893">770246e</a>)
• Update <a href="https://togithub.com/testing-library/dom">@testing-library/dom</a> (<a href="https://togithub.com/testing-library/react-testing-library/issues/931">#931</a>) (<a href="https://togithub.com/testing-library/react-testing-library/commit/05c74219cc19a46b0aa81a81db93121387bd4f8c">05c7421</a>)</p>
<p><strong>Features</strong></p>
<p>• Bump <a href="https://togithub.com/testing-library/dom">@testing-library/dom</a> (<a href="https://togithub.com/testing-library/react-testing-library/commit/6e6bf85301feed5d6bfa498d6d7f58a0bdef4520">6e6bf85</a>)
• Drop support for node 10 (<a href="https://togithub.com/testing-library/react-testing-library/issues/930">#930</a>) (<a href="https://togithub.com/testing-library/react-testing-library/commit/42dad789af193cdb11701db3fdb970f8ec0ffc83">42dad78</a>)</p>
<p><strong>BREAKING CHANGES</strong></p>
<p>• Bump <code>@testing-library/dom</code> to 8.0.0. Please check out the <a href="https://togithub.com/testing-library/dom-testing-library/releases/tag/v8.0.0"><code>@testing-library/dom@8.0.0</code> release page</a> for a detailed list of breaking changes.
• node 10 is no longer supported. It reached its end-of-life on 30.04.2021.</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-23 00:52:46
[cucumber/common] Comment on #1550
<p>> have separate repos in future though I agree that we should minimise the use of symlinks i</p>
<p>I agree.<br />
Why to use symlinks where relative path may be enough?</p>
<p>Symlinks may still be confusing sometimes when relative path are explicit anyway.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-23 01:46:56
[cucumber/common] Comment on #1605
<p><a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> thank you for your feedback.</p>
<p>Indeed I tried implementing the deserialization using reflection. I initially understood very well your proposition to add serialization/deserialization logic as part of the generated code.</p>
<p>I still would prefer to generate DTOs as light as possible for several reason.</p>
<p>Serialization is working very well with reflection, but yes, deserialization is very difficult that way, and there will always be issues with edge cases.</p>
<p>So, here's a new suggestion. What would you think of:</p>
<p>• generating lightweight DTOs with only the attributes
• using reflection for serialization as it is done in the current PR
• generating dedicated code for deserialization as part of the codegen script - as it is done in your PR - into a dedicated module outside of DTOs. Thanks to Ruby, it is super easy to reopen an existing class. To use the deserialization, a user would just have to opt-in for it requiring it explicitly.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-23 02:39:05
[cucumber/common] Comment on #1550
<p>> <a href="https://github.com/gasparnagy">@gasparnagy</a> try <a href="https://github.com/mattwynne/symlink-test">https://github.com/mattwynne/symlink-test</a> ?</p>
<p>Does not work. Git clone just clones it as normal file that contains the name of the linked file... :(</p>
<pre><code>&gt; type link-to-readme.md
README.md
</code></pre>
<p>So I think we can say that we should not use symlinks if we want to have contributors with Windows...</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-23 03:21:24
[cucumber/common] Comment on #1629
<p><strong>Branch automerge failure</strong></p>
<p>This PR was configured for branch automerge, however this is not possible so it has been raised as a PR instead.</p>
<hr />
<p>• Branch has one or more failed status checks</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-23 03:21:24
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
storybookjs/storybook
<strong><a href="https://togithub.com/storybookjs/storybook/blob/master/CHANGELOG.md#%E2%80%8B630-June-23-2021|<code">https://togithub.com/storybookjs/storybook/blob/master/CHANGELOG.md#%E2%80%8B630-June-23-2021|<code</a>v6.3.0</code>></strong></p>
<p><a href="https://togithub.com/storybookjs/storybook/compare/v6.2.9...v6.3.0">Compare Source</a></p>
<p><strong><a href="https://togithub.com/storybookjs/storybook/issues/14397">Optimized for UI development</a></strong></p>
<p>SB6.3 adds new UI development and testing features, while evolving with the JS ecosystem:</p>
<p>• 📐 Layout debugging with Measure and Outline addons
• 🔌 Reuse your stories in unit tests: Jest, Cypress & more
• 🚀 Frameworks: Angular 12 Ivy, Lit2 web components
• 🛠️ Builders: Webpack5 stable, Vite community
• 📦 Packaging: Modern ESM</p>
<p>It also contains hundreds more fixes, features, and tweaks. Browse the <a href="https://togithub.com/storybookjs/storybook/blob/next/CHANGELOG.md">changelogs</a> matching <code>6.3.0-alpha.**</code>, <code>6.3.0-beta.**</code>, and <code>6.3.0-rc.**</code> for the full list of changes. See <a href="https://storybook.js.org/blog/storybook-6-migration-guide/">Storybook 6 migration guide</a> to upgrade from <code>5.x</code> or <a href="https://togithub.com/storybookjs/storybook/blob/next/MIGRATION.md">MIGRATION.md</a> for detailed migration instructions.</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled due to failing status checks.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about these updates again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-23 07:07:14
[cucumber/common] Comment on #1625
<p>Lovely! Did you know we used to have a Texan translation at some point contributed by an idiot?</p>
<p><a href="https://github.com/cucumber-attic/gherkin2/commit/0e9322ae9e71c54590b062e4bedc812fd4c3b58c">cucumber-attic/gherkin2@0e9322a</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-23 07:35:21
[cucumber/common] Issue closed by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-23 07:36:18
[cucumber/common] Comment on #1623
<p>Hi <a href="https://github.com/sashokbg">@sashokbg</a> - I added an example in <a href="https://github.com/cucumber/common/tree/main/messages/java#readme">https://github.com/cucumber/common/tree/main/messages/java#readme</a> - hope this helps.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-23 09:00:59
[cucumber/common] Comment on #1625
<p>Hahaha that is too funny! I had a chuckle seeing that one!</p>
<p><#|…></p>
<p>On Wed, Jun 23, 2021, 9:07 AM Aslak Hellesøy <strong><em>*</strong>@<strong></em><em></strong>.<strong></em><em></strong>> wrote: Lovely! Did you know we used to have a Texan translation at some point contributed by an idiot? <strong></em><em></strong>@<strong></em><em></strong>.<strong></em>*</strong> <<a href="https://github.com/cucumber-attic/gherkin2/commit/0e9322ae9e71c54590b062e4bedc812fd4c3b58c">cucumber-attic/gherkin2@0e9322a</a>> — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <<a href="https://github.com/cucumber/common/pull/1625#issuecomment-866867645">#1625 (comment)</a>>, or unsubscribe <<a href="https://github.com/notifications/unsubscribe-auth/ADL4OVWMW4S2BAXEFCEUOEDTUHTBVANCNFSM47CZ3IQQ">https://github.com/notifications/unsubscribe-auth/ADL4OVWMW4S2BAXEFCEUOEDTUHTBVANCNFSM47CZ3IQQ</a>> .</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-23 10:01:37
[cucumber/common] Comment on #1550
<p>> Why to use symlinks where relative path may be enough?</p>
<p>Of course! I hadn't thought of that. Relative paths should work just fine.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-23 10:08:03
[cucumber/common] Comment on #1605
<p>IMO, if we want to keep the deserialization "clutter" out of the generated DTO classes, we'd be better off putting that code in a module and using <code>extend</code> to add it to the metaclass. That way the DTOs are simple to read, but the user doesn't have to think about requiring extra files to get the deserialization behaviour.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-23 10:11:07
[cucumber/common] Comment on #1625
<p>I need to see a full example of a scenario using this language! 😀</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-23 10:18:24
[cucumber/common] Comment on #1625
<p>> I need to see a full example of a scenario using this language! 😀</p>
<p>Haha, I'll try to get this for y'all! I need to run through y'all's checklist still. :D</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-23 10:21:53
[cucumber/common] Pull request opened by lamkr
<p><strong>Summary</strong></p>
<p>Implemented Gherkin in Dart language.</p>
<p><strong>Details</strong></p>
<p>The Gherkin project was ported to the Dart language.</p>
<p><strong>Motivation and Context</strong></p>
<p>I really like the Cucumber tool and I'm in love with Dart. Then I decided to combine the useful with the pleasant by implementing the tool in the language.</p>
<p><strong>How was this tested?</strong></p>
<p>I ported all tests in Java to the Dart project.<br />
I used Windows 10 Pro during implementation, and Dart 2.13.3 (stable) version.<br />
The change does not affect other areas of the code.</p>
<p><strong>Screenshots (if appropriate):</strong>
<strong>Types of changes</strong></p>
<p>• [] Bug fix (uninterrupted change that fixes a problem).
☑︎ New feature (uninterrupted change that adds functionality).
• [] Decisive change (fix or feature that would cause existing functionality not to work as expected).</p>
<p><strong>Checklist:</strong></p>
<p>• [] Change has been ported to Java.
• [] Change has been ported to Ruby.
• [] Change has been ported to JavaScript.
• [] Change has been moved to Go.
• [] Change has been ported to .NET.
☑︎ I added tests to my code.
☑︎ My change requires a change to the documentation.
☑︎ I updated the documentation accordingly.
• [] I updated CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-24 02:44:06
[cucumber/common] Comment on #1605
<p>> IMO, if we want to keep the deserialization "clutter" out of the generated DTO classes, we'd be better off putting that code in a module and using <code>extend</code> to add it to the metaclass. That way the DTOs are simple to read, but the user doesn't have to think about requiring extra files to get the deserialization behaviour.</p>
<p>Done.<br />
Deserialization are now part of messages per default.<br />
<code>require 'cucumber/messages'</code> now allows to use the DTOs, but also the deserializers.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-24 04:50:12
[cucumber/common] Comment on #1605
<p>> Serialization is working very well with reflection, but yes, deserialization is very difficult that way.</p>
<p>It's worse than difficult - it's impossible.</p>
<p>Knowledge of the schema is the only way we can infer the tight types to instantiate.</p>
<p>Good to see the deserialisation logic extracted to separate files! I would have preferred to do the same for the serialization logic (generate a separate file) rather than using reflection. Reflection adds performance overhead and complexity to the library, whereas with the statically generated version, there is no penalty for using reflection and regexp evaluation.</p>
<p>It's a tradeoff between complexity in the code generator and the runtime code, and I think I prefer to keep the complexity out of the runtime code.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-24 04:56:00
[cucumber/common] Comment on #1605
<p>What concrete use cases do we have for the <code>camelize</code> and <code>reject_nil_values</code> arguments?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-24 05:06:05
[cucumber/common] Comment on #1605
<p>Those are used for json serialization.<br />
<code>camelize</code> for the json using camelCase keys<br />
<code>reject_nil_values</code> to avoid serializing null values as requested here: <a href="https://github.com/cucumber/common/pull/1605#discussion_r653065795">#1605 (comment)</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-24 05:08:48
[cucumber/common] Comment on #1605
<p>I understand that, but why do we ever need to set <code>camelize</code> to <code>false</code> or <code>reject_nil_values</code> to <code>false</code>?</p>
<p>Was this flexibility added just in case someone else needs it? AFAICT we don't need it ourselves.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-24 05:12:30
[cucumber/common] Comment on #1605
<p>> I understand that, but why do we ever need to set <code>camelize</code> to <code>false</code> or <code>reject_nil_values</code> to <code>false</code>?
>
> Was this flexibility added just in case someone else needs it? AFAICT we don't need it ourselves.</p>
<p>Yes.</p>
<p>Because having a ruby hash with camelcase keys is not the ruby standard, and reject<em>nil</em>values to avoid data loss per default. So per default, the method has an intuitive use case.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-24 05:51:03
[cucumber/common] Pull request ready for review by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-24 05:53:10
[cucumber/common] Comment on #1605
<p>This speculative flexibility comes at a complexity and performance cost.</p>
<p>I assume you went with a reflection based serialization implementation in order to support this use case. If we implemented it without these option (always <code>camelize</code> and <code>reject_nil_values</code>), we wouldn't need the additional logic, and a static code generation would be sufficient (less complex and faster).</p>
<p>I think this is feature creep, but I'll rest my case here.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-24 06:29:28
[cucumber/common] Comment on #1605
<p>> This speculative flexibility comes at a complexity and performance cost.</p>
<p>The whole deserialization thing, which is the big piece in the current PR, is based on speculation. At first we wanted to implement only the serialization, which is required for the DTOs to work properly, then you requested the deserialization just in case someone else would need it, because it would be a more comprehensive developer experience. We actually do not need it ourselves.</p>
<p>The addition of a <code>to_h</code> method to messages is at first for our own usage. Make it intuitive and public for third-parties developer is a bonus. Actually those options do not add any complexity to the code as they result in only 2 extra <code>if</code> statement.</p>
<pre><code>def to_h(camelize: false, reject_nil_values: false)
resulting_hash = self.instance_variables.to_h do |variable_name|
h_key = variable_name[1..-1]
h_key = Cucumber::Messages::Message.camelize(h_key) if camelize
h_value = prepare_value(
self.instance_variable_get(variable_name),
camelize: camelize,
reject_nil_values: reject_nil_values
)
[ h_key.to_sym, h_value ]
end
resulting_hash.reject! { |_, value| value.nil? } if reject_nil_values
resulting_hash
end
def to_h
resulting_hash = self.instance_variables.to_h do |variable_name|
h_key = variable_name[1..-1]
h_key = Cucumber::Messages::Message.camelize(h_key)
h_value = prepare_value(
self.instance_variable_get(variable_name)
)
[ h_key.to_sym, h_value ]
end
resulting_hash.reject! { |_, value| value.nil? }
resulting_hash
end
</code></pre>
<p>> I assume you went with a reflection based serialization implementation in order to support this use case. If we implemented it without these option (always camelize and reject<em>nil</em>values), we wouldn't need the additional logic, and a static code generation would be sufficient (less complex and faster).</p>
<p>There is no relation between using reflection, and the addition of those two options. I would have proposed the same options with generated static code.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-24 06:56:27
[cucumber/common] Comment on #1605
<p>> It's a tradeoff between complexity in the code generator and the runtime code, and I think I prefer to keep the complexity out of the runtime code.</p>
<p>At the moment I stick to my idea for the serialization using reflection over code generation.</p>
<p>Indeed the runtime code may be a little bit more complex, but it has unit tests / specs that are easy to maintain, and I still find more easy to maintain something like this (that could still be improved in term of readibility):</p>
<pre><code>##
# Returns a new Hash formed from the message attributes
# If +camelize:+ keyword parameter is set to true, then keys will be camelized
# If +reject_nil_values:+ keyword parameter is set to true, resulting hash won't include nil values
#
# Cucumber::Messages::Duration.new(seconds: 1, nanos: 42).to_h # =&gt; { seconds: 1, nanos: 42 }
# Cucumber::Messages::PickleTag.new(name: 'foo', ast_node_id: 'abc-def').to_h(camelize: true) # =&gt; { name: 'foo', astNodeId: 'abc-def' }
# Cucumber::Messages::PickleTag.new(name: 'foo', ast_node_id: nil).to_h(reject_nil_values: true) # =&gt; { name: 'foo' }
#
# It is recursive so embedded messages are also processed
#
# location = Cucumber::Messages::Location.new(line: 2)
# Cucumber::Messages::Comment.new(location: location, text: 'comment').to_h # =&gt; { location: { line: 2, :column: nil }, text: "comment" }
#
def to_h(camelize: false, reject_nil_values: false)
resulting_hash = self.instance_variables.to_h do |variable_name|
h_key = variable_name[1..-1]
h_key = Cucumber::Messages::Message.camelize(h_key) if camelize
h_value = prepare_value(
self.instance_variable_get(variable_name),
camelize: camelize,
reject_nil_values: reject_nil_values
)
[ h_key.to_sym, h_value ]
end
resulting_hash.reject! { |_, value| value.nil? } if reject_nil_values
resulting_hash
end
</code></pre>
<p>Than something that would look like this:</p>
<pre><code>require 'cucumber/messages.dtos'
require 'json'
# The code was auto-generated by {this script}[<https://github.com/cucumber/common/blob/main/messages/jsonschema/scripts/codegen.rb>]
#
module Cucumber
module Messages
&lt;%- @schemas.sort.each do |key, schema| -%&gt;
class &lt;%= class_name(key) %&gt;
##
# Returns a new &lt;%= class_name(key) %&gt; from the given hash.
# If the hash keys are camelCased, they are properly assigned to the
# corresponding snake_cased attributes.
#
# Cucumber::Messages::&lt;%= class_name(key) %&gt;.from_h(some_hash) # =&gt; #&lt;Cucumber::Messages::&lt;%= class_name(key) %&gt;:0x... ...&gt;
#
def self.from_h(hash)
return nil if hash.nil?
self.new(
&lt;%-
schema['properties'].each do |property_name, property|
ref = property['$ref']
items_ref = property.dig('items', '$ref')
-%&gt;
&lt;%= "#{underscore(property_name)}: " -%&gt;
&lt;%- if items_ref -%&gt;hash[:&lt;%= property_name -%&gt;]&amp;.map { |item| &lt;%= class_name(items_ref) %&gt;.from_h(item) },
&lt;%- elsif ref -%&gt;&lt;%= class_name(ref) %&gt;.from_h(hash[:&lt;%= property_name %&gt;]),
&lt;%- else -%&gt;hash[:&lt;%= property_name %&gt;],
&lt;%- end -%&gt;
&lt;%- end -%&gt;
)
end
end
&lt;%- end -%&gt;
end
end
</code></pre>
<p>And regarding the performance overhead, at the moment cucumber-ruby is still based on an implementation of messages which embeds protobuf. I think the implementation here would already be an improvement.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-24 06:57:12
[cucumber/common] Comment on #1624
<p>This isn't scaling very well. For each different way to print we'd end up with a new method. Might be better to have the <code>TablePrinter</code> as part of the public API and deprecate the <code>print**</code> methods on the data table.</p>
<p>The table printer could take the <code>DataTable</code> as an argument and use <code>.cells()</code> to get the cells.</p>
<p>I can try to do this or you can if you want.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-24 12:13:56
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
webpack-contrib/style-loader
<strong><a href="https://togithub.com/webpack-contrib/style-loader/blob/master/CHANGELOG.md#%E2%80%8B300-httpsgithubcomwebpack-contribstyle-loadercomparev200v300-2021-06-24|<code">https://togithub.com/webpack-contrib/style-loader/blob/master/CHANGELOG.md#%E2%80%8B300-httpsgithubcomwebpack-contribstyle-loadercomparev200v300-2021-06-24|<code</a>v3.0.0</code>></strong></p>
<p><a href="https://togithub.com/webpack-contrib/style-loader/compare/v2.0.0...v3.0.0">Compare Source</a></p>
<p><strong>⚠️ BREAKING CHANGES</strong></p>
<p>• minimum supported <code>Node.js</code> version is <code>12.13.0</code>
• minimum supported <code>webpack</code> version is <code>5.0.0</code>
• the <code>modules.namedExport</code> option was removed, you don't need it anymore, because we respect the <code>modules.namedExport</code> option from <code>css-loader</code> (we just reexport all from <code>css-loader</code>), just remove it
• the <code>styleTag</code> value of the <code>injectType</code> (default value) option earlier uses singleton style tag by default for IE8-IE9 due limitations (<a href="https://www.telerik.com/blogs/internet-explorer-css-limits">more information</a>), in this release we have disabled this behavior, because these versions of IE are outdated, if you don't support these browsers this change does not affect you, if you require to support IE8-IE9, you can return old behaviour by setting <code>autoStyleTag</code> value for the <code>injectType</code> option (do the same for <code>lazyStyleTag</code>, i.e. change it to <code>lazyAutoStyleTag</code>)</p>
<p><strong>Features</strong></p>
<p>• added <code>autoStyleTag</code> and <code>lazyAutoStyleTag</code> values for the <code>injectType</code> option for compatibility of work modern and IE8-IE9 browsers
• added <code>styleTagTransform</code> option for custom processing style tags (useful if you need ponyfill CSS custom properties for IE8-IE10)
• reduce size of generated code
• reduce deps</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-24 12:52:28
[cucumber/common] Comment on #1630
<p>Hi <a href="https://github.com/lamkr">@lamkr</a> this is awesome! 🎉 thankyou so much! 🦄 🌈 ❤️ 💐</p>
<p>In terms of the build failures, I think it's down to us not yet having Dart on our <a href="https://github.com/cucumber/build/blob/main/Dockerfile">build container</a>.</p>
<p>Can you give me a bit of guidance as to what we need to add to that docker image so the new dart stuff will build?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-24 12:57:52
[cucumber/common] Comment on #1630
<p>If you want to chat 🍵 about this, you're welcome to join us in our <a href="https://cucumber.io/community#slack">community Slack</a> in the <a href="https://cucumberbdd.slack.com/archives/C62D0FK0E">#committers</a> channel 😄</p>
<p>I'm also trying to make myself available for live pairing ⚡ you can book a slot with me <a href="https://calendly.com/mattwynne">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-24 14:19:54
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-24 14:23:37
[cucumber/common] Comment on #1631
<p>Can you use the enum constant here instead of "FAILED"? (I'm sure there are other places in the code we should clean this up).</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-24 14:25:51
[cucumber/common] Comment on #1631
<p>Can we make it explicit that it's retries of <em>each scenario</em> here?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-24 14:33:24
[cucumber/common] aslakhellesoy requested changes on davidjgoss's pull request
<p>This is a breaking change to messages, so don't forget a changelog entry there.</p>
<p>Great PR!!</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 01:18:26
[cucumber/common] Comment on #1631
<p>Done</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 01:18:41
[cucumber/common] Comment on #1631
<p>Updated</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 01:21:59
[cucumber/common] Comment on #1631
<p>Side note: what do we think about using predictable ids and timing in <code>fake-cucumber</code> for this? I know we still wouldn't expect implementations to match exactly, but would at least stop every line in every fixture always changing.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 01:25:42
[cucumber/common] Pull request ready for review by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 01:28:35
[cucumber/common] Comment on #1631
<p>I put this in temporarily to get the build green, but just a little fuzzy on what I need to do to address it properly. Do I need to make changes to cucumber-ruby to add support code for the new CCK feature?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 01:46:11
[cucumber/common] Pull request ready for review by davidjgoss
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 01:47:16
[cucumber/common] Comment on #1631
<p>Yes, I think that's a good idea.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 01:49:52
[cucumber/common] Comment on #1631
<p>Okay - I'll make a separate PR for that once this lands.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 01:49:53
[cucumber/common] Comment on #1631
<p>Yes, I think that's what's needed.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 02:12:45
[cucumber/common] Comment on #1631
<p>At the moment, the ruby-testdata are based on an old version of cucumber-ruby.</p>
<p>In the PR <a href="https://github.com/cucumber/common/pull/1498">#1498</a> there is some changes to use current version of cucumber-ruby for those test data.<br />
It has not been merged yet because of the discussion around the empty scenarios still being in progress.</p>
<p>Maybe we could consider merging PR <a href="https://github.com/cucumber/common/pull/1498">#1498</a> - which add empty scenarios into the cck - and let implementations decide if they want to support it or not?</p>
<p>As I understand, supporting <code>retry</code> won't be supported by all implementation either?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 02:42:34
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 02:46:52
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
tj/commander.js
<strong><a href="https://togithub.com/tj/commander.js/blob/master/CHANGELOG.md#%E2%80%8B800-2021-06-25|<code">https://togithub.com/tj/commander.js/blob/master/CHANGELOG.md#%E2%80%8B800-2021-06-25|<code</a>v8.0.0</code>></strong></p>
<p><a href="https://togithub.com/tj/commander.js/compare/v7.2.0...v8.0.0">Compare Source</a></p>
<p><strong>Added</strong></p>
<p>• <code>.argument(name, description)</code> for adding command-arguments ([<a href="https://togithub.com/tj/commander.js/issues/1490">#1490</a>])
• supports default value for optional command-arguments ([<a href="https://togithub.com/tj/commander.js/issues/1508">#1508</a>])
• supports custom processing function ([<a href="https://togithub.com/tj/commander.js/issues/1508">#1508</a>])
• <code>.createArgument()</code> factory method ([<a href="https://togithub.com/tj/commander.js/issues/1497">#1497</a>])
• <code>.addArgument()</code> ([<a href="https://togithub.com/tj/commander.js/issues/1490">#1490</a>])
• <code>Argument</code> supports <code>.choices()</code> ([<a href="https://togithub.com/tj/commander.js/issues/1525">#1525</a>])
• <code>.showHelpAfterError()</code> to display full help or a custom message after an error ([<a href="https://togithub.com/tj/commander.js/issues/1534">#1534</a>])
• <code>.hook()</code> with support for <code>'preAction'</code> and <code>'postAction'</code> callbacks ([<a href="https://togithub.com/tj/commander.js/issues/1514">#1514</a>])
• client typing of <code>.opts()</code> return type using TypeScript generics ([<a href="https://togithub.com/tj/commander.js/issues/1539">#1539</a>])
• the number of command-arguments is checked for programs without an action handler ([<a href="https://togithub.com/tj/commander.js/issues/1502">#1502</a>])
• <code>.getOptionValue()</code> and <code>.setOptionValue()</code> ([<a href="https://togithub.com/tj/commander.js/issues/1521">#1521</a>])</p>
<p><strong>Changed</strong></p>
<p>• refactor and simplify TypeScript declarations (with no default export) ([<a href="https://togithub.com/tj/commander.js/issues/1520">#1520</a>])
• <code>.parseAsync()</code> is now declared as <code>async</code> ([<a href="https://togithub.com/tj/commander.js/issues/1513">#1513</a>])
• <em>Breaking:</em> <code>Help</code> method <code>.visibleArguments()</code> returns array of <code>Argument</code> ([<a href="https://togithub.com/tj/commander.js/issues/1490">#1490</a>])
• <em>Breaking:</em> Commander 8 requires Node.js 12 or higher ([<a href="https://togithub.com/tj/commander.js/issues/1500">#1500</a>])
• <em>Breaking:</em> <code>CommanderError</code> code <code>commander.invalidOptionArgument</code> renamed <code>commander.invalidArgument</code> ([<a href="https://togithub.com/tj/commander.js/issues/1508">#1508</a>])
• <em>Breaking:</em> TypeScript declaration for <code>.addTextHelp()</code> callback no longer allows result of <code>undefined</code>, now just <code>string</code> ([<a href="https://togithub.com/tj/commander.js/issues/1516">#1516</a>])
• refactor <code>index.tab</code> into a file per class ([<a href="https://togithub.com/tj/commander.js/issues/1522">#1522</a>])
• remove help suggestion from "unknown command" error message (see <code>.showHelpAfteError()</code>) ([<a href="https://togithub.com/tj/commander.js/issues/1534">#1534</a>])
• <code>Command</code> property <code>.arg</code> initialised to empty array (was previously undefined) ([<a href="https://togithub.com/tj/commander.js/issues/1529">#1529</a>])
• update dependencies</p>
<p><strong>Deprecated</strong></p>
<p>• second parameter of <code>cmd.description(desc, argDescriptions)</code> for adding argument descriptions ([<a href="https://togithub.com/tj/commander.js/issues/1490">#1490</a>])
• (use new <code>.argument(name, description)</code> instead)
• <code>InvalidOptionArgumentError</code> (replaced by <code>InvalidArgumentError</code>) ([<a href="https://togithub.com/tj/commander.js/issues/1508">#1508</a>])</p>
<p><strong>Removed</strong></p>
<p>• <em>Breaking:</em> TypeScript declaration for default export of global <code>Command</code> object ([<a href="https://togithub.com/tj/commander.js/issues/1520">#1520</a>])
• (still available as named <code>program</code> export)</p>
<p><strong>Migration Tips</strong></p>
<p>If you have a simple program without an action handler, you will now get an error if<br />
there are missing command-arguments.</p>
<pre><code>program
.option('-d, --debug')
.arguments('&lt;file&gt;');
program.parse();
</code></pre>
<pre><code>$ node trivial.js
error: missing required argument 'file'
</code></pre>
<p>If you want to show the help in this situation, you could check the arguments before parsing:</p>
<pre><code>if (process.argv.length === 2)
<a href="http://program.help">program.help</a>();
program.parse();
</code></pre>
<p>Or, you might choose to show the help after any user error:</p>
<pre><code>program.showHelpAfterError();
</code></pre>
<p><strong><a href="https://togithub.com/tj/commander.js/blob/master/CHANGELOG.md#%E2%80%8B720-2021-03-26|<code">https://togithub.com/tj/commander.js/blob/master/CHANGELOG.md#%E2%80%8B720-2021-03-26|<code</a>v7.2.0</code>></strong></p>
<p><a href="https://togithub.com/tj/commander.js/compare/v7.1.0...v7.2.0">Compare Source</a></p>
<p><strong>Added</strong></p>
<p>• TypeScript typing for <code>parent</code> property on <code>Command</code> ([<a href="https://togithub.com/tj/commander.js/issues/1475">#1475</a>])
• TypeScript typing for <code>.attributeName()</code> on <code>Option</code> ([<a href="https://togithub.com/tj/commander.js/issues/1483">#1483</a>])
• support information in package ([<a href="https://togithub.com/tj/commander.js/issues/1477">#1477</a>])</p>
<p><strong>Changed</strong></p>
<p>• improvements to error messages, README, and tests
• update dependencies</p>
<p><strong><a href="https://togithub.com/tj/commander.js/blob/master/CHANGELOG.md#%E2%80%8B710-2021-02-15|<code">https://togithub.com/tj/commander.js/blob/master/CHANGELOG.md#%E2%80%8B710-2021-02-15|<code</a>v7.1.0</code>></strong></p>
<p><a href="https://togithub.com/tj/commander.js/compare/v7.0.0...v7.1.0">Compare Source</a></p>
<p><strong>Added</strong></p>
<p>• support for named imports from ECMAScript modules ([<a href="https://togithub.com/tj/commander.js/issues/1440">#1440</a>])
• add <code>.cjs</code> to list of expected script file extensions ([<a href="https://togithub.com/tj/commander.js/issues/1449">#1449</a>])
• allow using option choices and variadic together ([<a href="https://togithub.com/tj/commander.js/issues/1454">#1454</a>])</p>
<p><strong>Fixed</strong></p>
<p>• replace use of deprecated <code>process.mainModule</code> ([<a href="https://togithub.com/tj/commander.js/issues/1448">#1448</a>])
• regression for legacy <code>command('**')</code> and call when command line includes options ([<a href="https://togithub.com/tj/commander.js/issues/1464">#1464</a>])
• regression for <code>on('command:**', ...)</code> and call when command line includes unknown options ([<a href="https://togithub.com/tj/commander.js/issues/1464">#1464</a>])
• display best error for combination of unknown command and unknown option (i.e. unknown command) ([<a href="https://togithub.com/tj/commander.js/issues/1464">#1464</a>])</p>
<p><strong>Changed</strong></p>
<p>• make TypeScript typings tests stricter ([<a href="https://togithub.com/tj/commander.js/issues/1453">#1453</a>])
• improvements to README and tests</p>
<p><strong><a href="https://togithub.com/tj/commander.js/blob/master/CHANGELOG.md#%E2%80%8B700-2021-01-15|<code">https://togithub.com/tj/commander.js/blob/master/CHANGELOG.md#%E2%80%8B700-2021-01-15|<code</a>v7.0.0</code>></strong></p>
<p><a href="https://togithub.com/tj/commander.js/compare/v6.2.1...v7.0.0">Compare Source</a></p>
<p><strong>Added</strong></p>
<p>• <code>.enablePositionalOptions()</code> to let program and subcommand reuse same option ([<a href="https://togithub.com/tj/commander.js/issues/1427">#1427</a>])
• <code>.passThroughOptions()</code> to pass options through to other programs without needing <code>--</code> ([<a href="https://togithub.com/tj/commander.js/issues/1427">#1427</a>])
• <code>.allowExcessArguments(false)</code> to show an error message if there are too many command-arguments on command line for the action handler ([<a href="https://togithub.com/tj/commander.js/issues/1409">#1409</a>])
• <code>.configureOutput()</code> to modify use of stdout and stderr or customise display of errors ([<a href="https://togithub.com/tj/commander.js/issues/1387">#1387</a>])
• use <code>.addHelpText()</code> to add text before or after the built-in help, for just current command or also for all subcommands ([<a href="https://togithub.com/tj/commander.js/issues/1296">#1296</a>])
• enhance Option class ([<a href="https://togithub.com/tj/commander.js/issues/1331">#1331</a>])
• allow hiding options from help
• allow restricting option arguments to a list of choices
• allow setting how default value is shown in help
• <code>.createOption()</code> to support subclassing of automatically created options (like <code>.createCommand()</code>) ([<a href="https://togithub.com/tj/commander.js/issues/1380">#1380</a>])
• refactor the code generating the help into a separate public Help class ([<a href="https://togithub.com/tj/commander.js/issues/1365">#1365</a>])
• support sorting subcommands and options in help
• support specifying wrap width (columns)
• allow subclassing Help class
• allow configuring Help class without subclassing</p>
<p><strong>Changed</strong></p>
<p>• <em>Breaking:</em> options are stored safely by default, not as properties on the command ([<a href="https://togithub.com/tj/commander.js/issues/1409">#1409</a>])
• this especially affects accessing options on program, use <code>program.opts()</code>
• revert behaviour with <code>.storeOptionsAsProperties()</code>
• <em>Breaking:</em> action handlers are p…</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 05:33:26
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 05:33:48
[cucumber/common] Comment on #1606
<p>Fixed by <a href="https://github.com/cucumber/common/pull/1632">#1632</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 05:33:48
[cucumber/common] Issue closed by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 07:55:30
[cucumber/common] Pull request ready for review by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 08:08:20
[cucumber/common] Comment on #249
<p>The PR <a href="https://github.com/cucumber/common/pull/1498">#1498</a> is ready for review.</p>
<p>It does not change any behavior in runtime code. It just add an acceptance tests of what is already happening.</p>
<p>Fake cucumber has been updated a little bit to match the behavior of cucumber-ruby because cucumber-ruby already has an explicit support for empty scenarios, cucumber-js has not - and also it was far more easy to do so due to how acceptance test-data are managed in the monorepo at the moment.</p>
<p>Also, I suggest to merge that PR not with the idea to impose the actual behavior neither to enforce the same behavior on all implementation, but to have some explicit spec for what is already happening now within the mono-repo.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 10:13:43
[cucumber/common] Comment on #1631
<p>Added the support code for ruby in cck and all green now.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 10:16:01
[cucumber/common] Comment on #1631
<p>> As I understand, supporting <code>retry</code> won't be supported by all implementation either?</p>
<p>Yep, just Ruby and JavaScript AFAIK. Other impls will just need to filter this one out of their CCK setup.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 11:48:44
[cucumber/common] Comment on #249
<p>The CCK is a canonical representation of how cucumber should work. Currently there is no agreement between implementations, currently there is also no consensus on what the correct behavior should be.</p>
<p>And I don't think a decision like this should be made without.</p>
<p>However adding a scenario to the cck, even without changing any implementation suggests that there is a correct behavior. Especially considering that in <a href="https://github.com/cucumber/common/pull/1498">#1498</a> you say:</p>
<p>> cucumber-ruby is already managing empty scenarios well so fake-cucumber has been updated in order to have the same behavior as cucumber-ruby.</p>
<p>So I don't really understand what is meant by:</p>
<p>> to have some explicit spec for what is already happening now within the mono-repo.</p>
<p>This seems to contravene the intended purpose of the CCK.</p>
<p>I am guessing there may be an underlying motivation that is rooted in the dual use of the CCK as a test set for the html formatter which may need an empty case for good measure. If this is the case it doesn't appear to have been stated.</p>
<p>And while I appreciate that you are being very diplomatic about the whole thing, logically either the CCK is no longer a CCK or we made a decision about the correct behavior of Cucumber without consensus.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 12:11:53
[cucumber/common] Comment on #1631
<p>Would be great to get this out in a patch release of cucumber-js so folks don't have to wait for 8.0.0</p>
<p>The fix to <code>query</code> will enable me to get a junit formatter done too.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 15:38:21
[cucumber/common] Comment on #1624
<p>Just pushed an update, if that's more what you had in mind? Am available next couple of days to fix it up further and clean up the documentation/tests etc, but equally my ego won't be bruised if you need to make some changes directly :)</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 15:42:00
[cucumber/common] Comment on #1624
<p>wasn't sure if your preference was to pass the table in the method, or when first calling the <code>TablePrinter()</code> constructor. I figured it's more likely this could be used to print multiple different tables with the same print configuration, so best to keep it as a method argument.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 15:43:34
[cucumber/common] Comment on #1624
<p>(also removed the printer-builder class from the last commit, and allowed for configuring the printer class directly, to minimise number of files, but I don't feel strongly about it)</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 23:50:34
[cucumber/common] Comment on #1624
<p>You can use <code>@link</code> to link to a method in Java doc.</p>
<p>Suggested change</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-25 23:51:07
[cucumber/common] Comment on #1624
<p>👍</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-26 00:03:06
[cucumber/common] Comment on #1624
<p>Getting close!</p>
<p>The idea is to use the builder pattern to create immutable table printer that can be used to print many different tables. Using the builder pattern to create immutable objects does take a bit more code but makes the objects much easier to work with. Especially in multi-threaded contexts.</p>
<p>So we have a <code>TablePrinter</code> with these public methods:</p>
<pre><code>TablePrinter
- static TablePrinter.Builder builder()
- printTable(DataTable table, StringBuilder appendable)
- printTable(DataTable table, Appendable appendable)
</code></pre>
<p>And a <code>TablePrinter.Builder</code> as a static inner class of <code>TablePrinter</code> with these public methods:</p>
<pre><code>TablePrinter.Builder
- indent(int)
- escape(boolean)
- TablePrinter build()
</code></pre>
<p>I realize this may not be apparent from the surrounding code but a lot of this code was written in a time before these ideas were common.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-26 00:03:45
[cucumber/common] Comment on #1624
<p>You've got it.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-26 13:02:06
[cucumber/common] Comment on #1624
<p>cool yeah, that makes sense.</p>
<p>Struggled how best to handle the child <code>DiffTablePrinter</code> class, whether it should have its own builder, or call the parent constructor with the parent's builder (for now went with the latter)</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-27 06:20:32
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-27 07:22:11
[cucumber/common] davidjgoss commented on hWorblehat's pull request
<p>Thanks for updating <a href="https://github.com/hWorblehat">@hWorblehat</a>.</p>
<p>Something that didn't really occur to me before was the disparity between what the user does with filters and how that's expressed in the URL.</p>
<p>For example, when the run includes statuses of PASSED, FAILED and UNDEFINED and we select:</p>
<p><a href="https://user-images.githubusercontent.com/3192745/123547954-c1992600-d75a-11eb-96b5-bc9311b84dab.png">image</a></p>
<p>In the URL we see:</p>
<pre><code>?hide=unknown&amp;hide=undefined
</code></pre>
<p>What would you think about something more like one of:</p>
<pre><code>?only=passed&amp;only=failed
?only=passed,failed
</code></pre>
<p>I think that would be more intuitive as it lines up with the filters UI, and make the special case for UNKNOWN an internal detail. Also, the filters wouldn't be reliant on which statuses were present in the test run - meaning for example you could have a bookmark to your latest build with <code>?only=failed</code> which would reliably show you only failures regardless of which other statuses were in play.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-27 07:49:23
[cucumber/common] Comment on #1624
<p>I had a look. Replaced inheritance with composition by extracting the <code>indent</code> as a function that returns a specific indent prefix for each row.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-27 08:10:50
[cucumber/common] mpkorstanje commented on artysidorenko's pull request
<p>I've rewritten the rendering logic from scratch to make the table printer thread-safe. Coincidently this also simplified the code a whole lot. Added a few tests to confirm I didn't break anything.</p>
<p>Thing I'm still wondering about is the naming. I'm not happy about the <code>Printer</code> suffix.</p>
<p>Comparable bits of code in the JDK are named <code>Formatter</code> and use the <code>format(Thing) : String</code> <code>formatTo(Thing, Appendable)</code>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-28 01:19:45
[cucumber/common] Comment on #249
<p>> The CCK is a canonical representation of how cucumber should work. Currently there is no agreement between implementations, currently there is also no consensus on what the correct behavior should be.
>
> And I don't think a decision like this should be made without.</p>
<p>Yes, that makes sense, actually 👍</p>
<p>> However adding a scenario to the cck, even without changing any implementation suggests that there is a correct behavior. Especially considering that in <a href="https://github.com/cucumber/common/pull/1498">#1498</a> you say:
>
> > cucumber-ruby is already managing empty scenarios well so fake-cucumber has been updated in order to have the same behavior as cucumber-ruby.
>
> So I don't really understand what is meant by:
>
> > to have some explicit spec for what is already happening now within the mono-repo.</p>
<p>I mean that no runtime code has been changed. Messages, Gherkin, no module in the monorepo that is then embedded into an implementation as runtime code is impacted by the PR</p>
<p>> This seems to contravene the intended purpose of the CCK.
>
> I am guessing there may be an underlying motivation that is rooted in the dual use of the CCK as a test set for the html formatter which may need an empty case for good measure. If this is the case it doesn't appear to have been stated.
>
> And while I appreciate that you are being very diplomatic about the whole thing, logically either the CCK is no longer a CCK or we made a decision about the correct behavior of Cucumber without consensus.</p>
<p>When I started working on that issue to move-on with it as it has been opened years ago, I did not realize there would be such a debate 😓 The CCK seemed to be a good starting point.</p>
<p>If the CCK intent is to specify the behavior of cucumber across implementation with no exception, given the discussion around empty scenarios, indeed it is actually not a good starting point 😅</p>
<p>Past week I asked for the review because empty scenarios are responsible for an issue in cucumber-js (<a href="https://github.com/cucumber/cucumber-js/issues/1668">cucumber/cucumber-js#1668</a>). Users are badly impacted by the non-support for that behavior. So I wanted to move-on in order to be able to fix the issue.</p>
<p>As I already mention: I think we may consider the possibility to have non-standard behavior depending implementations - based on the implementation ecosystem. I am not sure we may have a consensus on all and every topics like this one. As far as it does not impact the mono-repo, I think we should let us the possibility to have some <strong>explicit</strong> divergences over our implementations</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-28 01:25:58
[cucumber/common] Comment on #249
<p><a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> FWIW I'm not convinced it's causing that serious a cucumber-js issue (see my <a href="https://github.com/cucumber/cucumber-js/issues/1668#issuecomment-853962610">latest comment there</a>). Still it would be great to somehow find a way forward with this!</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-28 02:11:00
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-28 03:11:11
[cucumber/common] Comment on #249
<p>Would it be possible to add "individual" test cases to the formatters? That way we don't have to abuse the CCK because we want to test a few edge cases?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-28 05:14:02
[cucumber/common] Comment on #249
<p>That sounds interesting 👍</p>
<p>On the other hand, "abusing" the CCK - but properly - would keep all our spec - even the one that diverge based on the implementation - at the same place.</p>
<p>I am still confused with the CCK. Do we consider it as</p>
<p>• the place to specify cucumber behaviors?
• the place to specify <strong>the common</strong> cucumber behaviors?
• a non-reg test suite?</p>
<p>I know you said that it is a "canonical representation of how cucumber should work". But I am not sure to understand what do you mean here - and it would be the same with the french translation of that sentence: I am not sure what "canonical" means 😅</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-28 05:55:29
[cucumber/common] Comment on #1624
<p>Looks good, thanks!</p>
<p>The only difference I noticed is that the comparable <code>java.util.Formatter</code> tends to keep the appendable as a property of the instance, but I think the <code>formatTo</code> method makes the behaviour pretty clear here 👍</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-28 07:55:37
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-29 00:12:57
[cucumber/common] aurelien-reeves approved davidjgoss's pull request
<p>Looks good to me 👍</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-29 00:16:30
[cucumber/common] Pull request merged by aurelien-reeves
<https://github.com/cucumber/common/pull/1628|#1628 Update dependency @testing-library/react to v12>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-29 00:41:39
[cucumber/common] Comment on #1581
<p>We declare a lot of node images: 10, 12, 14, 15, but actually we do not use them all.</p>
<p>Often the 12 is used.<br />
10 is used once with cucumber-expressions, which is also build using node 12<br />
15 is used once with html-formatter, which is built only with that one</p>
<p>Should we:</p>
<p>• use the current LTS candidate (18) and remove the other ones
• use the oldest LTS (12), the current LTS candidate (18) for all javascript jobs, and remove the other ones
• use all the LTS (12, 14, 16, 18) for all javascript jobs
• use all the active versions (12, 14, 16, 17, 18) for all javascript jobs
• something else?</p>
<p>Currently except for cucumber-expressions, each javascript module is build only once using node-js 12. Adding more node version will make the CI far more difficult to maintain as-is and may need a proper refactoring that could be part of a dedicated PR to avoid renovate messing with potential active developments.</p>
<p>Note: with Ruby, we also declare a lot of docker images with different ruby version.<br />
All ruby gems are built using the latest one except for gherkin which is built with each ruby version declared. May we consider doing the same with node?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-29 01:14:56
[cucumber/common] Comment on #1581
<p>I pushed an update that copy the ruby behavior: use the active node version for all js packages, and also build gherkin with all currently active node versions.</p>
<p>What do you think?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-29 01:16:29
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-29 01:26:11
[cucumber/common] Comment on #1581
<p>It appears there are no official docker images for node 17 and 18 at the moment.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-29 01:47:32
[cucumber/common] Comment on #714
<p>We have a major release of cucumber-js coming up. Thoughts on deprecating <code>--no-strict</code> and then removing it next time?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-29 02:16:59
[cucumber/common] Comment on #1631
<p>Thanks <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a>!</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-29 02:17:29
[cucumber/common] Comment on #1631
<p><a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> you okay with this one now?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-29 05:14:23
[cucumber/common] Comment on #714
<p>(or even just removing it now...)</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-29 05:34:29
[cucumber/common] Comment on #714
<p>I would prefer deprecating it first</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-29 05:38:54
[cucumber/common] Comment on #1581
<p>The issue with cucumber-react is due to tests using react. Mocha hang when tests are using react components.</p>
<p>Here's a minimal reproducible example:</p>
<pre><code>// react/javascript/test/test.tsx
import ReactDOM from 'react-dom'
import React from 'react'
import { JSDOM } from 'jsdom'
describe('mocha hanging issue', () =&gt; {
it('Minimal reproducible example', () =&gt; {
const dom = new JSDOM('&lt;html lang="en"&gt;&lt;body&gt;&lt;div id="content"&gt;&lt;/div&gt;&lt;/body&gt;&lt;/html&gt;')
const container = dom.window.document.getElementById('content')
const app = &lt;div&gt;&lt;/div&gt;
ReactDOM.render(app, container)
})
})
</code></pre>
<pre><code>react/javascript$ npx mocha test/test.tsx
</code></pre>
<p>Using <code>act</code> from react testing libraries behaves the same.<br />
I did not found anything on the Internet regarding an issue like that.</p>
<p><a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> <a href="https://github.com/davidjgoss">@davidjgoss</a> <a href="https://github.com/charlierudolph">@charlierudolph</a> any idea?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-29 06:58:13
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-29 06:59:12
[cucumber/common] Comment on #1581
<p><a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> I'd suggest we do our primary builds with the newest LTS version which is 14. Maybe we can get that in and then circle back to this mocha issue? 16 is still in a somewhat unstable phase I think, so issues like this could shake out in patches.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-29 07:20:35
[cucumber/common] Comment on #1581
<p>> <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> I'd suggest we do our primary builds with the newest LTS version which is 14. Maybe we can get that in and then circle back to this mocha issue? 16 is still in a somewhat unstable phase I think, so issues like this could shake out in patches.</p>
<p>I'll give the v14 a try.<br />
Node really has a weird life cycle IMO 🤔</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-29 17:33:26
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-29 23:23:47
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-30 00:36:49
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-06-30 12:17:36
[cucumber/common] Pull request opened by renovate[bot]
<https://github.com/cucumber/common/pull/1637|#1637 Update dependency mini-css-extract-plugin to v2>
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
webpack-contrib/mini-css-extract-plugin
<strong><a href="https://togithub.com/webpack-contrib/mini-css-extract-plugin/blob/master/CHANGELOG.md#%E2%80%8B200-httpsgithubcomwebpack-contribmini-css-extract-plugincomparev162v200-2021-06-30|<code">https://togithub.com/webpack-contrib/mini-css-extract-plugin/blob/master/CHANGELOG.md#%E2%80%8B200-httpsgithubcomwebpack-contribmini-css-extract-plugincomparev162v200-2021-06-30|<code</a>v2.0.0</code>></strong></p>
<p><a href="https://togithub.com/webpack-contrib/mini-css-extract-plugin/compare/v1.6.2...v2.0.0">Compare Source</a></p>
<p><strong>NOTES</strong></p>
<p>In the current release we have fixed many problems with <code>publicPath</code>, previously to generate relative URLs inside CSS files developers use different hacks: <code>publicPath: ''</code>, <code>publicPath: '../'</code>, using relative <code>../../</code> in the <code>outputPath</code> option for <code>file-loader</code> and etc. Now you don't need it anymore. Webpack v5 uses <code>publicPath: "auto"</code> by default, which means to always generate relative URLs, and now <code>mini-css-extract-plugin</code> does the same.</p>
<p><strong>We strongly recommend use <code>auto</code> value by default (except when using CDN).</strong></p>
<p><strong>We also want to show you that the <a href="https://togithub.com/webpack-contrib/file-loader"><code>file-loader</code></a> and <a href="https://togithub.com/webpack-contrib/url-loader"><code>url-loader</code></a> are outdated, please migrate on <a href="https://webpack.js.org/guides/asset-modules/">Asset Modules</a>.</strong></p>
<p><strong>⚠️ BREAKING CHANGES</strong></p>
<p>• minimum supported <code>Node.js</code> version is <code>12.13.0</code>
• minimum supported <code>webpack</code> version is <code>5.0.0</code>
• the <code>modules.namedExport</code> option was removed, you don't need it anymore, because we respect the <code>modules.namedExport</code> option from <code>css-loader</code>, just remove it
• the <code>publicPath</code> option no longer automatically adds <code>/</code> (trailing slash), you need to specify, you need to specify it yourself if it is absent, i.e. if you have <code>publicPath: "/my/public/path"</code> replace it with <code>publicPath: "/my/public/path/"</code></p>
<p><strong>Bug Fixes</strong></p>
<p>• generating correct relative <code>url()</code> for assets inside CSS files when you use <code>publicPath: "auto"</code> (default value)</p>
<p><strong><a href="https://togithub.com/webpack-contrib/mini-css-extract-plugin/compare/v1.6.1...v1.6.2">1.6.2</a> (2021-06-28)</strong>
<strong>Bug Fixes</strong></p>
<p>• performance improvement</p>
<p><strong><a href="https://togithub.com/webpack-contrib/mini-css-extract-plugin/compare/v1.6.0...v1.6.1">1.6.1</a> (2021-06-25)</strong>
<strong>Bug Fixes</strong></p>
<p>• memory leaks (<a href="https://togithub.com/webpack-contrib/mini-css-extract-plugin/commit/c68aca7c58b1bfb3d3f9f8db70fd814e50f82aa1">c68aca7</a>)</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 00:43:12
[cucumber/common] Pull request merged by aurelien-reeves
<https://github.com/cucumber/common/pull/1637|#1637 Update dependency mini-css-extract-plugin to v2>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 01:52:46
[cucumber/common] Comment on #1605
<p>I'll get onto this properly on friday afternoon. So I can dedicate a bit of time to it.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 05:28:14
[cucumber/common] Comment on #871
<p>First pass at a JavaScript implementation here: <a href="https://github.com/davidjgoss/cucumberjs-junit-formatter">https://github.com/davidjgoss/cucumberjs-junit-formatter</a></p>
<p>Here's an example from a test of what it outputs: <a href="https://github.com/davidjgoss/cucumberjs-junit-formatter/blob/main/test/snapshots/integration.test.ts.md">https://github.com/davidjgoss/cucumberjs-junit-formatter/blob/main/test/snapshots/integration.test.ts.md</a></p>
<p>It's wrapped as a cucumber-js <code>Formatter</code> class, but doesn't use any of the old formatter helpers - just consumes envelopes and uses gherkin query and cucumber query.</p>
<p>We're now talking about having it live in the monorepo as a generic envelopes -> junit stream with a thin binding in cucumber-js (similar to the <code>html</code> formatter). We could then port other platforms' junit formatters here and share a test suite driven from the compatibility kit (which would ideally need a few more cases e.g. pending, undefined, ambiguous).</p>
<p>Things to maybe iron out:</p>
<p>• As <a href="https://github.com/mpkorstanje">@mpkorstanje</a> points out, for each <code>&lt;testcase&gt;</code> the combination of <code>classname</code> and <code>name</code> should be unique. Currently these are populated with <code>pickle.uri</code> and <code>pickle.name</code> respectively, but the latter isn't guaranteed to be unique (e.g. with scenario outlines) so need to add some kind of qualifier.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 05:41:22
[cucumber/common] Comment on #871
<p>The code that makes the names unique can be found here:</p>
<p><a href="https://github.com/cucumber/cucumber-jvm/blob/main/core/src/main/java/io/cucumber/core/plugin/JUnitFormatter.java#L208">https://github.com/cucumber/cucumber-jvm/blob/main/core/src/main/java/io/cucumber/core/plugin/JUnitFormatter.java#L208</a></p>
<p>But probably best not to copy that as it assumes scenarios are executed in order. This is not guaranteed for messages.</p>
<p>If not unique, I would suggest numbering the nodes in a feature file and using the path to the pickle as a unique prefix.</p>
<pre><code>1. Rule
1.1 Scenario
1.1.1 Examples Section
1.1.1.1 Example 1
1.1.1.2 Example 2
1.1.1.3 Example 3
1.1.2 Examples Section
1.1.2.1 Example 1
1.1.2.2 Example 2
1.1.2.3 Example 3
2. Rule
2.1 Scenario
....
</code></pre>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 05:42:49
[cucumber/common] Comment on #871
<p><a href="https://github.com/mpkorstanje">@mpkorstanje</a> if doing that wouldn't it be easier to use line number? As that's already in the AST right? Or am I getting confused?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 06:49:40
[cucumber/common] Comment on #871
<p>Yes. Would be easier, but the reason why it is there would be less obvious.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 06:52:56
[cucumber/common] Comment on #871
<p>> As <a href="https://github.com/mpkorstanje">@mpkorstanje</a> points out, for each <code>&lt;testcase&gt;</code> the combination of <code>classname</code> and <code>name</code> should be unique. Currently these are populated with <code>pickle.uri</code> and <code>pickle.name</code> respectively, but the latter isn't guaranteed to be unique (e.g. with scenario outlines) so need to add some kind of qualifier.</p>
<p>If it is the combination of <code>classname</code> and <code>name</code> which should be unique, then <code>pickle.uri</code> and <code>pickle.name</code> should work. Am I missing something?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 07:00:26
[cucumber/common] Comment on #871
<p><a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> pickles in the same feature file could have the same name if:</p>
<p>• You have a scenario outline with examples (unless the name includes a parameter)
• People have just given several scenarios exactly the same name (bad, not invalid Gherkin)</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 07:04:26
[cucumber/common] Comment on #871
<p>Oh, ok.</p>
<p>What would you think of adding the line number in the URI, as an anchor?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 07:07:01
[cucumber/common] Comment on #871
<p>Okay. Other suggestion then. We can do what ever JUnit does when it encounters duplicate.</p>
<p>I've seen JUnit do something like this (I think, would have to double check)</p>
<pre><code>ExampleTest
ParameterizedTest [0]
ParameterizedTest [1]
ParameterizedTest [2]
ParameterizedTest [3]
</code></pre>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 07:07:43
[cucumber/common] Comment on #871
<p>> <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> pickles in the same feature file could have the same name if:
>
> • You have a scenario outline with examples (unless the name includes a parameter)
> • People have just given several scenarios exactly the same name (bad, not invalid Gherkin)</p>
<p>I have the following gherkin frequently (Which I don't consider bad)</p>
<pre><code>Feature: Foo
Rule: English language
Scenario: Can do Foo
Rule: Welsh language
Scenario: Can do Foo
</code></pre>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 07:26:06
[cucumber/common] Comment on #871
<p><a href="https://github.com/luke-hill">@luke-hill</a> true - I forget that the rule name doesn't contribute to the pickle name</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 07:26:15
[cucumber/common] Pull request merged by mpkorstanje
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 07:26:16
[cucumber/common] Comment on #1624
<p>Hi <a href="https://github.com/artysidorenko">@artysidorenko</a>,</p>
<p>Thanks for your making your first contribution to Cucumber, and welcome to the Cucumber committers team! You can now push directly to this repo and all other repos under the cucumber organization! 🍾</p>
<p>In return for this generous offer we hope you will:</p>
<p>• ✅ Continue to use branches and pull requests. When someone on the core team approves a pull request (yours or someone else's), you're welcome to merge it yourself.
• 💚 Commit to setting a good example by following and upholding our <a href="https://cucumber.io/conduct">code of conduct</a> in your interactions with other collaborators and users.
• 💬 Join the <a href="http://cucumber.io/support#slack">community Slack channel</a> to meet the rest of the team and make yourself at home.
• ℹ️ Don't feel obliged to help, just do what you can if you have the time and the energy.
• 🙋 Ask if you need anything. We're looking for feedback about how to make the project more welcoming, so please tell us!</p>
<p>On behalf of the Cucumber core team,<br />
Aslak Hellesøy<br />
Creator of Cucumber</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 07:26:53
[cucumber/common] Comment on #1624
<p>I'll have to find some time to make the release.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 07:35:22
[cucumber/common] aslakhellesoy approved davidjgoss's pull request
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 07:35:29
[cucumber/common] Issue closed by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 07:35:30
[cucumber/common] Issue closed by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 07:35:31
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 07:36:45
[cucumber/common] Issue opened by aurelien-reeves
<p>Refs. <a href="https://github.com/cucumber/cucumber-js/issues/1453">cucumber/cucumber-js#1453</a></p>
<p><strong>Is your feature request related to a problem? Please describe.</strong></p>
<p>The idea came from the discussion in an issue opened for cucumber-js: <a href="https://github.com/cucumber/cucumber-js/issues/1453">cucumber/cucumber-js#1453</a><br />
It may help third-party developers to have more info regarding failures in the message testStepFinished.</p>
<p>Info which have been identified as nice to have:</p>
<p>• expected
• actual
• error message</p>
<p>Would those info be easy to retrieve from our various implementation?</p>
<p>The issue on cucumber-js also request for the error object itself, but that may be more difficult (impossible?) to properly serialize it in a message.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 07:40:29
[cucumber/common] Comment on #871
<p>> Okay. Other suggestion then. We can do what ever JUnit does when it encounters duplicate.
>
> I've seen JUnit do something like this (I think, would have to double check)
>
> <code>
&gt; ExampleTest
&gt; ParameterizedTest [0]
&gt; ParameterizedTest [1]
&gt; ParameterizedTest [2]
&gt; ParameterizedTest [3]
&gt;
&gt;</code></p>
<p>Only for duplicates right?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 07:41:46
[cucumber/common] Comment on #1605
<p>> The whole deserialization thing, which is the big piece in the current PR, is based on speculation. At first we wanted to implement only the serialization, which is required for the DTOs to work properly, then you requested the deserialization just in case someone else would need it, because it would be a more comprehensive developer experience. We actually do not need it ourselves.</p>
<p>Fair point. I think we should publish a library that will enable people to write consumers in Ruby. See <a href="https://github.com/cucumber/common/issues/1623">#1623</a> where someone was asking how to consume messages in Java.</p>
<p>It's speculative, but it also encouragement.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 07:42:06
[cucumber/common] aslakhellesoy approved aurelien-reeves's pull request
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 07:42:23
[cucumber/common] Comment on #1638
<p>Perhaps also a stacktrace?</p>
<p>For Java this is relatively easy to support this for testing libraries that use opentest4j.</p>
<p><a href="https://github.com/ota4j-team/opentest4j/blob/master/src/main/java/org/opentest4j/AssertionFailedError.java">https://github.com/ota4j-team/opentest4j/blob/master/src/main/java/org/opentest4j/AssertionFailedError.java</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 07:45:19
[cucumber/common] Comment on #1605
<p>Thank you <a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> 🙂</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 07:46:58
[cucumber/common] Comment on #1638
<p>I though about it a stacktrace too, yes.<br />
Extract it from the message may be an interesting improvement and simplify working with failed steps later</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 08:14:10
[cucumber/common] Comment on #871
<p><a href="https://github.com/mpkorstanje">@mpkorstanje</a> <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> thoughts on omitting the <code>[0]</code> for the first one? Would make it a bit simpler to implement I think.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 09:20:35
[cucumber/common] Comment on #871
<p>> Only for duplicates right?</p>
<p>Yep! Here's a PR that does that <a href="https://github.com/davidjgoss/cucumberjs-junit-formatter/pull/1">davidjgoss/cucumberjs-junit-formatter#1</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 10:40:49
[cucumber/common] Pull request opened by davidjgoss
<p><strong>Summary</strong></p>
<p>For generating the golden master for compatibility-kit, use <code>--predictable-ids</code> when calling fake-cucumber.</p>
<p><strong>Motivation and Context</strong></p>
<p>Every time we add/change something in the compatibility kit, every line of every file in the golden master changes because of the ids and timestamps - this makes it difficult to see the actual changes and also to find any meaning in the git history.</p>
<p>We wouldn't expect Cucumber implementations to match these when validating against the golden master messages - it's just about reducing the churn in the source.</p>
<p><strong>Checklist:</strong></p>
<p>☐ The change has been ported to Java.
☐ The change has been ported to Ruby.
☐ The change has been ported to JavaScript.
☐ The change has been ported to Go.
☐ The change has been ported to .NET.
☐ I've added tests for my code.
☐ My change requires a change to the documentation.
☐ I have updated the documentation accordingly.
☐ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 10:43:30
[cucumber/common] Comment on #1631
<p><a href="https://github.com/cucumber/common/pull/1639">#1639</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-01 14:10:25
[cucumber/common] Comment on #871
<p>Nasty case:</p>
<pre><code>Scenario: A
Scenario: A
Scenario: A[1]
</code></pre>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-02 01:30:18
[cucumber/common] Comment on #1533
<p><a href="https://github.com/thiagojacinto">@thiagojacinto</a> the message formatter only reports status at the step level and doesn't aggregate/summarise - this is down to consuming tools. You'll see all the test step results for every attempt, so when there have been retries you probably just want to disregard the data from previous attempts. This is what we're <a href="https://github.com/cucumber/common/pull/1631/files#diff-a8181df4dd2488263c17d5ced829886e8d02029f851a374e84f8e3032256ccb7">doing now in <code>@cucumber/query</code></a>, using the <code>testCaseStarted</code> message as a signal to clear previous results if they exist.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-02 04:12:22
[cucumber/common] Comment on #1302
<p>Hi <a href="https://github.com/davidjgoss">@davidjgoss</a>,</p>
<p>The problem generally with the status filters is that the UI only unambiguously reflects the filtering behaviour for the current set of results. For instance, if we have a set of results with only <code>PASSED</code> and <code>FAILED</code> statuses, then we only have check boxes for those 2 statuses. What should the filtering behaviour be for the <code>SKIPPED</code> status? For these results, it doesn't matter, but we need to make a decision about how it translates into URL query parameters, as that same URL may refer to a different set of results in the future.</p>
<p>Originally, I went with the <code>hide=</code> scheme, as I felt this had the most explicit relationship with textboxes: if a checkbox is shown and unchecked, then you get a <code>hide=</code> parameter, and if a <code>hide</code> URL parameter exists for a status that you don't get a checkbox for, that's because that status doesn't appear in the results, so it doesn't matter. Also, it has the advantage that you won't accidentally hide a status that just happens not to be present in the current set of results by toggling check-boxes and copying the URL (you have to explicitly untick a checkbox for each status you want to hide). It feels safer to construct a URL that by default <em>shows</em> statuses that the user hasn't explicitly set a check-box for. However, I do agree that it's jarring, particularly WRT the unknown status (more on that below).</p>
<p>Currently, I have added a change that uses your <code>only=</code> suggestion:</p>
<p>• If all shown checkboxes are ticked, there is no query parameter and no filtering, even if the URL later points to a result set that includes more statuses.
• If at least one shown checkbox is unticked, then there will be a query parameters and filtering to only show those statuses for which there <em>are</em> ticked checkboxes.
• A URL with a single <code>only=</code> parameter with nothing after the <code>=</code> corresponds to all checkboxes being unticked, and no results being shown.
• If the URL includes an <code>only=</code> parameter for a status that isn't present in the current set of results, then the filtering will still be applied accordingly, but as usual there will be no checkbox for that status. As soon as the user changes any of the other checkboxes in the UI, that URL parameter will be removed, changing the URL to match the current state of the checkboxes only.</p>
<p>That last point in particular can be quite jarring, particularly if the set of results only includes a single status, but not the one you have an <code>only=</code> filter for. In that case, there are no checkboxes (old behaviour, which I've kept, only includes checkboxes when there is more than one status present), so the only way to alter the filtering is to manually change the URL. To avoid this, we could change the UI to always include checkboxes for status filters, even if there's only one – we couldn't do it conditionally on whether there was a superfluous filter in the URL, because as soon as that checkbox was interacted with, it would disappear as the URL is brought in line with it!</p>
<p>I implemented the <code>only=</code> scheme because initially I thought I agreed it would be better. Having thought about it more, particularly writing this response, I've now changed my mind again. I think the <code>hide=</code> scheme is cleaner, more explicit, and safer in terms of not accidentally excluding results. Still, since I've implemented <code>only=</code> now, I'll wait for your response before reverting.</p>
<p>Finally, regarding <code>unknown</code>. My original reading of the old code was that <code>unknown</code> statuses are always silently filtered out. Hence their annoying presence in the <code>hide=</code> filters as soon as you did something non-default. However, looking again I think I was wrong. It looks like <code>unknown</code> wasn't treated specially for filtering, except that if it was the only status present, then a checkbox would still appear for it in the search bar, giving the user the ability to filter it out (thus showing no results). Unless my reading of the old code is wrong, I can't understand why we would want this behaviour at all, so my current proposal would be to remove all special handling for <code>unknown</code>, whichever URL scheme we choose. Am I missing something here?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-02 06:51:25
[cucumber/common] Comment on #568
<p>From SUSE, we are at least using it in SUSE Manager and Uyuni projects.<br />
This is the project <a href="https://www.uyuni-project.org/">https://www.uyuni-project.org/</a> and this is the code <a href="https://github.com/uyuni-project/uyuni/tree/master/testsuite">https://github.com/uyuni-project/uyuni/tree/master/testsuite</a><br />
You can find the Uyuni logo from there.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-02 07:58:55
[cucumber/common] Comment on #871
<p><a href="https://github.com/mpkorstanje">@mpkorstanje</a> oh come on, that's just gratuitous! 😆</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-02 08:56:14
[cucumber/common] aslakhellesoy approved davidjgoss's pull request
<p>Thanks <a href="https://github.com/davidjgoss">@davidjgoss</a> - a simple and very useful improvement!</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-02 11:16:26
[cucumber/common] Pull request merged by davidjgoss
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-03 03:37:34
[cucumber/common] Pull request opened by mpkorstanje
<p>Writing JSON directly to stream avoids the overhead of writing it to a string<br />
first.</p>
<p>☐ Bug fix (non-breaking change which fixes an issue).
☐ New feature (non-breaking change which adds functionality).
☑︎ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<p><strong>Checklist:</strong></p>
<p>☑︎ The change has been ported to Java.
☑︎ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-03 03:40:09
[cucumber/common] Comment on #1624
<p>Done. Unfortunately there are some breaking changes that precede your change. If you want to continue this in Cucumber JVM. I would suggest branching of <a href="https://github.com/cucumber/cucumber-jvm/commit/c985807332024cf1c46d59d16c0641c983744d2d">this commit</a> in the <code>v7.x.x</code> branch. Not everything may work, there is a bit of a re-factoring coming through.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-03 09:39:35
[cucumber/common] Comment on #1624
<p>Nevermind that. Branch of <code>main</code>. v7 is too broken.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-05 00:32:31
[cucumber/common] Comment on #1302
<p>I must say that the <code>?hide=</code> approach does not bother me.<br />
If feels like 'I am opting-in to hide some specific statuses as they are all visible per default'.</p>
<p>Per default, all checkboxes are activated and the user uncheck some to filter the result, and that seems to properly reflect with the URI, as we can see in the short clip.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-05 01:22:01
[cucumber/common] Comment on #1617
<p><a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> do you want to keep it opened for further discussion?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-05 01:23:24
[cucumber/common] Comment on #1603
<p>Once <a href="https://github.com/cucumber/common/pull/1605">#1605</a> is merged, I'll be able to continue on this one. Hopefully there won't be too much to do to make it pass.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-05 01:54:06
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-05 02:07:06
[cucumber/common] Comment on #1302
<p>Good points <a href="https://github.com/hWorblehat">@hWorblehat</a> <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a></p>
<p>> I implemented the only= scheme because initially I thought I agreed it would be better. Having thought about it more, particularly writing this response, I've now changed my mind again. I think the hide= scheme is cleaner, more explicit, and safer in terms of not accidentally excluding results. Still, since I've implemented only= now, I'll wait for your response before reverting.</p>
<p>Sure, let's revert if it looks like it's not an improvement. Really appreciate you giving it a run though!</p>
<p>> Finally, regarding unknown. My original reading of the old code was that unknown statuses are always silently filtered out. Hence their annoying presence in the hide= filters as soon as you did something non-default. However, looking again I think I was wrong. It looks like unknown wasn't treated specially for filtering, except that if it was the only status present, then a checkbox would still appear for it in the search bar, giving the user the ability to filter it out (thus showing no results). Unless my reading of the old code is wrong, I can't understand why we would want this behaviour at all, so my current proposal would be to remove all special handling for unknown, whichever URL scheme we choose. Am I missing something here?</p>
<p>No, I agree, let's remove the treatment of unknown. Thanks!</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-05 03:51:44
[cucumber/common] Comment on #1302
<p>Reverted back to using <code>hide=</code>. Awaiting final approval...</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-05 05:20:53
[cucumber/common] Comment on #1617
<p>> do you want to keep it opened for further discussion?</p>
<p>No, I think it's good to merge now</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-05 06:57:15
[cucumber/common] Comment on #1617
<p><a href="https://github.com/cucumber/common/pull/1605">#1605</a> has been merged<br />
It will be difficult to merge both PR</p>
<p>My question was more "do we keep it opened or do we close it" 😅</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-05 07:24:53
[cucumber/common] Comment on #1617
<p>Ah my bad, I was mixing up the two issues. Closing this one.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-05 07:24:55
[cucumber/common] Pull request closed by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-05 09:02:02
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
webpack/webpack
<strong><a href="https://togithub.com/webpack/webpack/releases/v5.42.1"><code>v5.42.1</code></a></strong></p>
<p><a href="https://togithub.com/webpack/webpack/compare/v5.42.0...v5.42.1">Compare Source</a></p>
<p><strong>Bugfixes</strong></p>
<p>• fix crashes when rebuilding with <code>jsonData</code> or <code>dataUrl</code> of undefined</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled due to failing status checks.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-05 09:02:15
[cucumber/common] Comment on #1641
<p><strong>Branch automerge failure</strong></p>
<p>This PR was configured for branch automerge, however this is not possible so it has been raised as a PR instead.</p>
<hr />
<p>• Branch has one or more failed status checks</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-05 13:48:07
[cucumber/common] Pull request merged by renovate[bot]
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 00:08:03
[cucumber/common] Comment on #1302
<p>The need for the <code>--exit</code> option indicates we may have some async work still waiting in the background.<br />
Did you need this even before adding your own tests?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 00:08:03
[cucumber/common] Comment on #1302
<p>Suggested change</p>
<p>I would be more user centric here. What do you think?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 00:08:04
[cucumber/common] aurelien-reeves commented on hWorblehat's pull request
<p>It looks pretty good to me.<br />
But I would prefer <a href="https://github.com/davidjgoss">@davidjgoss</a> <a href="https://github.com/charlierudolph">@charlierudolph</a> or <a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> to review it too. They are more experienced with that package than I am.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 01:12:09
[cucumber/common] Comment on #1579
<p><a href="https://github.com/l3l0">@l3l0</a> do you have the possibility to add the second translation as suggested by <a href="https://github.com/jakzal">@jakzal</a>?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 01:41:28
[cucumber/common] Comment on #1559
<p>Hi <a href="https://github.com/enkessler">@enkessler</a></p>
<p>There will be a new update soon removing the use of the plain ruby hashes and using DTOs instead - among other things.</p>
<p>Those DTOs should provide a more reliable API in the future.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 02:09:31
[cucumber/common] Comment on #1302
<p>Removing <code>--exit</code> didn't cause an issue for me locally. Just pushed a commit to remove it, let's see what happens in CI.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 03:23:25
[cucumber/common] Comment on #1302
<p>Without it mocha was hanging for me (when running via <code>make</code>). Must be the version of Node I'm using or something else. TBH I didn't investigate whether it was code/tests that I'd added that was causing the hang.</p>
<p>I had a lot of problems generally with building. For instance, after changing code in the react library, running <code>PACKAGES="messages react html-formatter" LANGUAGES=javascript make</code> didn't seem to update html-formatter accordingly. I then tried running <code>make clean</code> first, but that caused the subsequent build to fail. In the end, my rebuild cycle was:</p>
<pre><code>git clean -dfX
PACKAGES="messages react html-formatter" LANGUAGES=javascript make
</code></pre>
<p>which is somewhat annoying as it meant I was reinstalling NPM packages every time. Dunno what I was doing wrong, if anything.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 03:25:22
[cucumber/common] Comment on #1302
<p>Also, I couldn't find it documented anywhere, but trial and error suggests that the build system specifically requires NPM v7.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 04:22:47
[cucumber/common] Comment on #1302
<p>> Also, I couldn't find it documented anywhere, but trial and error suggests that the build system specifically requires NPM v7.</p>
<p>You're right and we do need to document that. Those changes happened after you first raised this PR but the devex is quite a bit simpler for JS as a result. From the root of the project you can run <code>npm i &amp;&amp; npm run build</code> and then you should just be able to run tests etc from any module without any extra building/linking/installing.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 04:23:12
[cucumber/common] davidjgoss approved hWorblehat's pull request
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 04:23:20
[cucumber/common] Pull request merged by davidjgoss
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 04:26:53
[cucumber/common] Comment on #1302
<p>👍 Thanks guys</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 05:40:32
[cucumber/common] Comment on #1302
<p>Thank you <a href="https://github.com/hWorblehat">@hWorblehat</a>!</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 06:31:27
[cucumber/common] Comment on #1640
<p>Is it ready to be merged? Or something else remains to be done?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 07:03:44
[cucumber/common] Comment on #1640
<p>I think I need to fix the version numbers. I think <code>make post-release</code> was forgotten.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 09:49:41
[cucumber/common] Comment on #1630
<p>Hello, <a href="https://github.com/mattwynne">@mattwynne</a> ! Thank you by your help and sorry for the delay in my reply!<br />
I tested the commands below here on my machine and they worked. For us to be sure, could you test it there too, please?</p>
<pre><code># Install Dart
ENV DART_SDK /usr/lib/dart
RUN apt-get install apt-transport-https
RUN sh -c 'curl -s -L <https://dl-ssl.google.com/linux/linux_signing_key.pub> | apt-key add -'
RUN sh -c 'curl -s -L <https://storage.googleapis.com/download.dartlang.org/linux/debian/dart_stable.list> \
-o /etc/apt/sources.list.d/dart_stable.list'
RUN apt-get update
RUN apt-get install dart
ENV PATH $DART_SDK/bin:$PATH
</code></pre>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 13:47:28
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 13:47:37
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 14:01:58
[cucumber/common] Pull request merged by mpkorstanje
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 14:02:22
[cucumber/common] Pull request merged by mpkorstanje
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 14:28:38
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 14:29:25
[cucumber/common] Comment on #1630
<p>Thanks <a href="https://github.com/lamkr">@lamkr</a>. I tried it out locally on my M1 Mac and I had some problems. See <a href="https://github.com/cucumber/build/pull/36">cucumber/build#36</a></p>
<p>I think it might be due to me having an arm64 architecture machine. Let's discuss it over <a href="https://github.com/cucumber/build/pull/36">there</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 14:47:01
[cucumber/common] Pull request merged by mpkorstanje
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 18:23:18
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled due to failing status checks.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 18:23:32
[cucumber/common] Comment on #1645
<p><strong>Branch automerge failure</strong></p>
<p>This PR was configured for branch automerge, however this is not possible so it has been raised as a PR instead.</p>
<hr />
<p>• Branch has one or more failed status checks</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 20:54:40
[cucumber/common] Comment on #1302
<p>Thanks for working with us on this <a href="https://github.com/hWorblehat">@hWorblehat</a> - it's a great feature for the HTML report to have!</p>
<p>Thanks to you too <a href="https://github.com/davidjgoss">@davidjgoss</a> for keeping the feedback rolling!</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-06 23:31:52
[cucumber/common] Pull request merged by renovate[bot]
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 02:47:37
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 02:47:37
[cucumber/common] Comment on #1625
<p>Hi <a href="https://github.com/willmac321">@willmac321</a>,</p>
<p>Thanks for your making your first contribution to Cucumber, and welcome to the Cucumber committers team! You can now push directly to this repo and all other repos under the cucumber organization! 🍾</p>
<p>In return for this generous offer we hope you will:</p>
<p>• ✅ Continue to use branches and pull requests. When someone on the core team approves a pull request (yours or someone else's), you're welcome to merge it yourself.
• 💚 Commit to setting a good example by following and upholding our <a href="https://cucumber.io/conduct">code of conduct</a> in your interactions with other collaborators and users.
• 💬 Join the <a href="http://cucumber.io/support#slack">community Slack channel</a> to meet the rest of the team and make yourself at home.
• ℹ️ Don't feel obliged to help, just do what you can if you have the time and the energy.
• 🙋 Ask if you need anything. We're looking for feedback about how to make the project more welcoming, so please tell us!</p>
<p>On behalf of the Cucumber core team,<br />
Aslak Hellesøy<br />
Creator of Cucumber</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 02:53:36
[cucumber/common] Issue opened by aurelien-reeves
<p>The serial build is slow. It may last for about 50 minutes.</p>
<p>I don't like to merge PR before all jobs are done. Sometimes it is difficult because of changes on the main branches which need to be merged in the PR, then waiting for the serial build to finish and hoping no other changes would be done on main in the meantime...</p>
<p>I know, I have the possibility to force merge thanks to my admin authorization. But this remain a workaround. If we could prevent it, that would be great.</p>
<p>So I suggest to try speeding-up the CI. Any ideas?</p>
<p>My first idea would be to split that serial build into languages-specific builds. That way we would have parallel builds per modules, and parallel builds per languages.</p>
<p>Would we have specific requirements that would not be met with such parallelization?</p>
<p>Would you have other ideas?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 03:18:18
[cucumber/common] Comment on #1579
<p><a href="https://github.com/l3l0">@l3l0</a> <a href="https://github.com/jakzal">@jakzal</a> I've made the update. Does it looks good?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 03:40:39
[cucumber/common] Comment on #1646
<p>There is an implicit build order encoded in the make file. We re-encode that order in our parallel build through dependencies. When the make file order is wrong, we can't build locally and this has broken often in the past.</p>
<p>If we de remove the serial build, we do need something to confirm that building locally will still work.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 04:03:57
[cucumber/common] Comment on #1579
<p><a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> looks good to me 👍</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 05:05:17
[cucumber/common] Comment on #1646
<p>I understand the need for the serial build. I suggest to split it per languages: one build for all java modules, one for javascript, ...</p>
<p>I think that it should cover the need you described?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 05:05:36
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 05:05:36
[cucumber/common] Comment on #1579
<p>Hi <a href="https://github.com/l3l0">@l3l0</a>,</p>
<p>Thanks for your making your first contribution to Cucumber, and welcome to the Cucumber committers team! You can now push directly to this repo and all other repos under the cucumber organization! 🍾</p>
<p>In return for this generous offer we hope you will:</p>
<p>• ✅ Continue to use branches and pull requests. When someone on the core team approves a pull request (yours or someone else's), you're welcome to merge it yourself.
• 💚 Commit to setting a good example by following and upholding our <a href="https://cucumber.io/conduct">code of conduct</a> in your interactions with other collaborators and users.
• 💬 Join the <a href="http://cucumber.io/support#slack">community Slack channel</a> to meet the rest of the team and make yourself at home.
• ℹ️ Don't feel obliged to help, just do what you can if you have the time and the energy.
• 🙋 Ask if you need anything. We're looking for feedback about how to make the project more welcoming, so please tell us!</p>
<p>On behalf of the Cucumber core team,<br />
Aslak Hellesøy<br />
Creator of Cucumber</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 05:11:04
[cucumber/common] Comment on #1603
<p>If no objection soon, I'll force-merge it to be able to start releasing new versions of some modules in the mono-repo</p>
<p>If any objection, I'll wait for proper reviews :)</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 05:20:26
[cucumber/common] Comment on #1640
<p>All <code>pom.xml</code> are looking good to me</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 05:20:43
[cucumber/common] Comment on #1646
<p>I think there's some cross-language dependency in places - e.g. the java and ruby versions of <code>html-formatter</code> need the javascript one to have built.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 05:26:47
[cucumber/common] Comment on #1498
<p>Back to draft as there is no consensus regarding empty scenarios</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 05:29:31
[cucumber/common] Comment on #1646
<p>Thanks for the info</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 05:42:05
[cucumber/common] aslakhellesoy approved aurelien-reeves's pull request
<p>LGTM!</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 05:58:05
[cucumber/common] Pull request merged by mpkorstanje
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 06:01:55
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 10:05:17
[cucumber/common] Pull request opened by renovate[bot]
<https://github.com/cucumber/common/pull/1647|#1647 Update module github.com/onsi/gomega to v1.14.0>
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
onsi/gomega
<strong><a href="https://togithub.com/onsi/gomega/releases/v1.14.0"><code>v1.14.0</code></a></strong></p>
<p><a href="https://togithub.com/onsi/gomega/compare/v1.13.0...v1.14.0">Compare Source</a></p>
<p><strong>1.14.0</strong>
<strong>Features</strong></p>
<p>• gmeasure.SamplingConfig now suppers a MinSamplingInterval [<a href="https://togithub.com/onsi/gomega/commit/e94dbca"><code>e94dbca</code></a>]
• Eventually and Consistently support functions that make assertions [<a href="https://togithub.com/onsi/gomega/commit/2f04e6e"><code>2f04e6e</code></a>]
• Eventually and Consistently now allow their passed-in functions to make assertions.<br />
These assertions must pass or the function is considered to have failed and is retried.
• Eventually and Consistently can now take functions with no return values. These implicitly return nil<br />
if they contain no failed assertion. Otherwise they return an error wrapping the first assertion failure. This allows<br />
these functions to be used with the Succeed() matcher.
• Introduce InterceptGomegaFailure - an analogue to InterceptGomegaFailures - that captures the first assertion failure<br />
and halts execution in its passed-in callback.</p>
<p><strong>Fixes</strong></p>
<p>• Call Verify GHTTPWithGomega receiver funcs (<a href="https://togithub.com/onsi/gomega/issues/454">#454</a>) [<a href="https://togithub.com/onsi/gomega/commit/496e6fd"><code>496e6fd</code></a>]
• Build a binary with an expected name (<a href="https://togithub.com/onsi/gomega/issues/446">#446</a>) [<a href="https://togithub.com/onsi/gomega/commit/7356360"><code>7356360</code></a>]</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 10:05:17
[cucumber/common] Comment on #1647
<p><strong>⚠️ Artifact update problem</strong></p>
<p>Renovate failed to update an artifact related to this branch. You probably do not want to merge this PR as-is.</p>
<p>♻️ Renovate will retry this branch, including artifacts, only when one of the following happens:</p>
<p>• any of the package files in this branch needs updating, or
• the branch becomes conflicted, or
• you check the rebase/retry checkbox if found above, or
• you rename this PR's title to start with "rebase!" to trigger it manually</p>
<p>The artifact failure details are included below:</p>
<p><strong>File name: json-formatter/go/go.sum</strong></p>
<pre><code>Command failed: docker run --rm --name=renovate_go --label=renovate_child -v "/mnt/renovate/gh/cucumber/common":"/mnt/renovate/gh/cucumber/common" -v "/tmp/renovate-cache":"/tmp/renovate-cache" -v "/tmp/renovate-cache/others/go":"/tmp/renovate-cache/others/go" -e GOPATH -e CGO_ENABLED -w "/mnt/renovate/gh/cucumber/common/json-formatter/go" <a href="http://docker.io/renovate/go:1.16.5">docker.io/renovate/go:1.16.5</a> bash -l -c "git config --global url.\"https://****redacted****@github.com/\".insteadOf \"<https://github.com/\>" &amp;&amp; go get -d ./..."
go: <a href="http://github.com/cucumber/common/messages/go/v16@v16.0.1">github.com/cucumber/common/messages/go/v16@v16.0.1</a>: reading <a href="http://github.com/cucumber/common/messages/go/messages/go/go.mod">github.com/cucumber/common/messages/go/messages/go/go.mod</a> at revision messages/go/v16.0.1: unknown revision messages/go/v16.0.1
</code></pre>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 10:53:52
[cucumber/common] Issue closed by mattwynne
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 10:54:09
[cucumber/common] Issue closed by mattwynne
<https://github.com/cucumber/common/issues/1532|#1532 Remove mono dependency from the build process>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 10:57:52
[cucumber/common] Comment on #1646
<p><a href="https://github.com/cucumber/build/issues/18">cucumber/build#18</a> is somewhat related to this.</p>
<p><a href="https://github.com/ciaranmcnulty">@ciaranmcnulty</a> and I have had one pairing session on it, so it's too early to say what impact it could have.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-07 23:51:07
[cucumber/common] Comment on #1646
<p>> <a href="https://github.com/cucumber/build/issues/18">cucumber/build#18</a> is somewhat related to this.
>
> <a href="https://github.com/ciaranmcnulty">@ciaranmcnulty</a> and I have had one pairing session on it, so it's too early to say what impact it could have.</p>
<p>I am not sure that would be related.<br />
I think the serial build is slow because we build everything in every language without any parallelization.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-08 02:21:55
[cucumber/common] Pull request opened by aurelien-reeves
<p>@types/node should match the version of nodejs we are using</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-08 02:26:00
[cucumber/common] davidjgoss approved aurelien-reeves's pull request
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-08 02:26:33
[cucumber/common] Comment on #1648
<p>I just want to wait for the CI to be green before merging it. To be sure it is actually ok for real 😅</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-08 05:04:56
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-08 05:17:52
[cucumber/common] Issue opened by aurelien-reeves
<p><a href="https://github.com/cucumber/common/blob/main/messages/ruby/cucumber-messages.gemspec#L9">https://github.com/cucumber/common/blob/main/messages/ruby/cucumber-messages.gemspec#L9</a></p>
<p>It still mention protocol buffers but now, we are providing DTOs generated based on json schema.</p>
<p>As I am in release process at the moment, I open that issue as a reminder.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-08 05:24:48
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
cucumber/messages-java
<strong><a href="https://togithub.com/cucumber/messages-java/compare/v16.0.1...v17.0.0"><code>v17.0.0</code></a></strong></p>
<p><a href="https://togithub.com/cucumber/messages-java/compare/v16.0.1...v17.0.0">Compare Source</a></p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about these updates again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-08 05:25:52
[cucumber/common] Comment on commit "Post release messages v3.0.0"
<p>It was the post release for messages 17.0.0, not 3.0.0, sorry for the typo</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-08 06:27:25
[cucumber/common] Issue opened by aurelien-reeves
<p>Since we do want to use the monorepo for go modules, we have to add dedicated tags.</p>
<p>For example for messages v17.0.0, the tag <code>messages/go/v17.0.0</code> has been created.</p>
<p>The creation of such tags has to be part of the release scripts</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-08 07:02:38
[cucumber/common] New release Release - cucumber-gherkin/v20.0.0 published by cukebot
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-08 08:15:53
[cucumber/common] Comment on #1651
<p>The following should fix this: <a href="https://github.com/cucumber/common/commit/ac27f50bfcc3aaf8558c3e63ce00c53cae1f5d39">ac27f50</a></p>
<p>Anyone to review it? I pushed this on main directly because there are no CI for that anyway.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-08 10:38:48
[cucumber/common] Comment on #1559
<p>Huzzah for stability! Thanks!</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-08 11:01:09
[cucumber/common] Comment on #1651
<p>Well spotted. Looks good to me.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-08 12:15:12
[cucumber/common] New release Release - json-formatter/go/v19.0.0 published by cukebot
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-08 12:36:16
[cucumber/common] Pull request closed by renovate[bot]
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-08 23:47:44
[cucumber/common] Pull request merged by aurelien-reeves
<https://github.com/cucumber/common/pull/1647|#1647 Update module github.com/onsi/gomega to v1.14.0>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-08 23:48:24
[cucumber/common] Issue closed by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-09 01:58:42
[cucumber/common] Comment on #1574
<p>I close this one as messages v17.0.0 brings brand new ruby DTOs :)</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-09 01:58:42
[cucumber/common] Issue closed by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-09 11:03:07
[cucumber/common] Comment on #1574
<p>🎉</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-09 13:49:06
[cucumber/common] Comment on #1646
<p>I came over this today - something to consider?</p>
<p><a href="https://earthly.dev/">https://earthly.dev/</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-10 15:53:10
[cucumber/common] Pull request opened by aslakhellesoy
<p><strong>Summary</strong></p>
<p>Use <a href="https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/RegExp/exec">RegExpExecArray#indices</a> to get the position of capture groups.</p>
<p><strong>Details</strong></p>
<p>This PR removes the dependency on the <a href="https://www.npmjs.com/package/becke-ch--regex--s0-0-v1--base--pl--lib">becke-ch--regex--s0-0-v1--base--pl--lib</a> library and uses <code>RegExpExecArray#indices</code> instead.</p>
<p><a href="https://github.com/tc39/proposal-regexp-match-indices">RegExp Match Indices</a> were introduced in v8 v9.0.259. The first Node.js release to use a v8 newer than that is <a href="https://nodejs.org/en/download/releases/">16.4.0</a>.</p>
<p><em>THIS SHOULD NOT BE MERGED UNTIL WE DROP SUPPORT FOR NODEJS < 16.4.0</em> (which is probably in 2023/2024).</p>
<p>Alternatively we could merge this sooner if we use the <a href="https://github.com/rbuckton/regexp-match-indices">polyfill</a>.</p>
<p><strong>Motivation and Context</strong></p>
<p>The main motivation for this change is to stop depending on <a href="http://www--s0-v1.becke.ch/tool/becke-ch--regex--s0-v1/becke-ch--regex--s0-0-v1--homepage--pl--client/">http://www--s0-v1.becke.ch/tool/becke-ch--regex--s0-v1/becke-ch--regex--s0-0-v1--homepage--pl--client/</a>, which seems to be a liability to depend on because it doesn't seem to have a public Git repo, except for <a href="https://github.com/TheWorldTable/regexp-plus">this fork</a>.</p>
<p>I would also expect this to be slightly more performant since it's based on a native implementation, but I doubt it will be significant for end users.</p>
<p>Also see <a href="https://github.com/cucumber/common/issues/300">#300</a></p>
<p><strong>How Has This Been Tested?</strong></p>
<p>The tests pass on Node v16.4.2, but I doubt it will pass on Node < 16.4.0 as explained above.</p>
<p><strong>Types of changes</strong></p>
<p>☐ Bug fix (non-breaking change which fixes an issue).
☐ New feature (non-breaking change which adds functionality).
☑︎ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<p><strong>Checklist:</strong></p>
<p>☐ The change has been ported to Java.
☐ The change has been ported to Ruby.
☑︎ The change has been ported to JavaScript.
☐ The change has been ported to Go.
☐ The change has been ported to .NET.
☐ I've added tests for my code.
☐ My change requires a change to the documentation.
☐ I have updated the documentation accordingly.
☐ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-10 16:07:05
[cucumber/common] Comment on #1646
<p>Looks very interesting, potentially a general improvement.</p>
<p>Its parallism doesn't appear to be distributable across multiple nodes. So maybe not a solution to ensure the local build keeps working.</p>
<p>However if the commands are simple enough, we may not have to.</p>
<p>--</p>
<p>Does come to mind that we did setup a system for node that builds all js modules at once. So that also breaks the modular nature of the monorepo.</p>
<p>We can't say that the modules build in parallel are the same as those in serial. It is a different build process.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-10 16:14:02
[cucumber/common] Comment on #1652
<p>Update: I added the polyfill so we don't have to wait until 2023/2024 to merge this.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-12 01:53:37
[cucumber/common] aurelien-reeves approved aslakhellesoy's pull request
<p>👍</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-12 14:38:38
[cucumber/common] Comment on #1652
<p>👍</p>
<p>If we're using the polyfill let's edit the text in the PR to reflect that and remove the warnings.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-12 15:56:54
[cucumber/common] Pull request ready for review by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-12 16:22:26
[cucumber/common] Comment on #1652
<p>PR text updated</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-12 16:22:45
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-12 23:43:57
[cucumber/common] Comment on #1653
<p><a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> I have written a very basic Gherkin autocomplete for the eclipse-java plugin based on the eclipse-api, so an official API would be great, even in the form of some rules that could be parsed and translated to some other techniques.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-12 23:44:54
[cucumber/common] Comment on #1653
<p>This is also related to <a href="https://github.com/cucumber/common/issues/1306">#1306</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-13 00:51:56
[cucumber/common] Comment on #1653
<p>What does the eclipse plugin use for suggestions <a href="https://github.com/laeubi">@laeubi</a>?</p>
<p>The other Gherkin auto-complete plugins I have seen suggest regular/cucumber expressions from step definitions that the user has to edit into a step afterward. Does the Eclipse plugin do the same thing? If so, I think a Java port of this library could improve suggestions considerably for Eclipse.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-13 06:35:35
[cucumber/common] Issue opened by karthrav
<p><strong>Describe the bug</strong><br />
Line number 3 i.e. import { version } from '../package.json'; throws a compile error TS2614: Module '"<em>.json"' has no exported member 'version'. Did you mean to use 'import version from "</em>.json"' instead?</p>
<p>This occurs only when you use "tsc --resolveJsonModule false" compile command.<br />
Issue does not occur if you use simply tsc.</p>
<p><strong>To Reproduce</strong><br />
Steps to reproduce the behavior:</p>
<ol><li>Compile your project with tsc --resolveJsonModule false</li><li>See error</li><li>Compile your project with tsc</li><li>No error</li>
</ol>
<p><strong>Expected behavior</strong><br />
No error should occur when compiling with tsc --resolveJsonModule false</p>
<p><strong>Screenshots</strong><br />
<a href="https://user-images.githubusercontent.com/87373079/125452193-4308a756-adc3-4c86-ae94-a98df5af797b.png">image</a><br />
<a href="https://user-images.githubusercontent.com/87373079/125453237-f32a03f3-e39b-43f2-87fd-a5ada792a69a.png">image</a></p>
<p><strong>Desktop:</strong></p>
<p>• OS: Windows</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-13 06:35:44
[cucumber/common] Pull request opened by aurelien-reeves
<https://github.com/cucumber/common/pull/1655|#1655 Convert ruby messages duration.nanos to integer>
<p><code>duration.nanos</code> is described in the schema as being an integer. It was already the case with protobuf.</p>
<p>With the new DTOs, <code>duration.nanos</code> became a float. When using the json-formatter to convert ndjson to "legacy" json, an error is raised:</p>
<pre><code>ERROR: json: cannot unmarshal number 127100.0 into Go struct field Duration.testStepFinished.testStepResult.duration.nanos of type int64
</code></pre>
<p>Converting nanos to integer should fix this.</p>
<p><strong>How Has This Been Tested?</strong></p>
<p>As durations are removed when validating messages as part of the neutralize-json scripts, this is not properly tested yet in the mono-repo.<br />
I encounter this issue while trying to reactivate the CCK tests with cucumber-ruby. So there will be at least that way to test it.</p>
<p><strong>Types of changes</strong></p>
<p>☑︎ Bug fix (non-breaking change which fixes an issue).</p>
<p><strong>Checklist:</strong></p>
<p>☑︎ I've added tests for my code.
☑︎ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-13 06:52:20
[cucumber/common] Pull request ready for review by aurelien-reeves
<https://github.com/cucumber/common/pull/1655|#1655 Convert ruby messages duration.nanos to integer>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-13 07:17:17
[cucumber/common] Comment on #1655
<p>I think we should also patch json-formatter to handle floats, so it works with those ruby versions that output a float.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-13 07:23:44
[cucumber/common] Comment on #1655
<p>> I think we should also patch json-formatter to handle floats, so it works with those ruby versions that output a float.</p>
<p>For now no ruby version output float at the moment.<br />
However I check how difficult that would be</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-13 07:25:38
[cucumber/common] Comment on #1655
<p>Ok I misunderstood. The go code chokes on a float string. Where does that come from?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-13 07:28:31
[cucumber/common] Comment on #1655
<p>From some tests I did on cucumber-ruby main branch but not released yet</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-13 08:02:13
[cucumber/common] Comment on #1655
<p>> I think we should also patch json-formatter to handle floats, so it works with those ruby versions that output a float.</p>
<p>Due to the strong type system of Golang, that may be doable but it looks pretty difficult.<br />
As the issue has not been released yet, would it be ok to merge and release this PR quickly in order to avoid releasing this known bug with cucumber-ruby? That way, we would not have to patch json-formatter</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-13 08:25:23
[cucumber/common] Comment on #1653
<p>I'm on vacation, so just a sort feedback: I like the concept. Two things came to my mind based on my auto-complete implementations:</p>
<ol><li>For SpecFlow we would need to also record whether the particular step definition can be used as Given, When, Then or all -- but this can be added later.</li><li>In large codebases the parameter variations can be too big. So I would apply some maximum threshold for the possible parameter values (like take the top 10).</li>
</ol>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-13 11:08:12
[cucumber/common] Comment on #1654
<p>We could avoid requiring consuming projects to use <code>--resolveJsonModule</code> by swapping the <code>import</code> out for a <code>require</code> (and probably suppressing the eslint error). It compiles down to a <code>require</code> either way, but the different is in the <code>.d.ts</code> output.</p>
<p>Out of interest <a href="https://github.com/karthrav">@karthrav</a> is there a particular reason you'd prefer to avoid <code>--resolveJsonModule</code>? JSON imports are being standardised in JavaScript so it seems a reasonable default IMO (not sure if TypeScript would ever change it though).</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-13 11:35:30
[cucumber/common] Comment on #1653
<p>Thanks for the feedback <a href="https://github.com/gasparnagy">@gasparnagy</a>! I’ll add the max thing. I might as well add the GWT bit too.</p>
<p>unless there are objections I’ll merge and release 0.0.1 of this tonight. Dying to try it out on some real data (in Cucumber Reports).</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-13 13:36:45
[cucumber/common] Comment on #1653
<p>This looks fantastic!</p>
<p>When handling parameters (e.g. <code>${2|belly,suitcase,table|}</code>), does the LSP allow us to express constraints on custom values? We have the regex from the parameter type, so theoretically that could allow the editor validate a custom value, if the LSP allows for that.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-13 16:10:32
[cucumber/common] Comment on #1653
<p>> This looks fantastic!</p>
<p>Much obliged :-)</p>
<p>> When handling parameters (e.g. ${2|belly,suitcase,table|}), does the LSP allow us to express constraints on custom values?</p>
<p>I don't think the LSP allows constraints to be defined on choices, nor do I think it is a good idea. An editor should never <em>prevent</em> users from typing something, even if it is wrong. That would be a patronising and annoying user experience.</p>
<p>Instead, it should <em>highlight</em> that you've typed something invalid, and this is what editors do, using syntax validation. So if the user doesn't want <code>42</code> or <code>54</code> (from the <code>${1|42,54|}</code> parameter), they should be allowed to press <code>ESC</code> and insert <code>horse</code>, even if the underlying parameter is a Cucumber Expression <code>{int}</code> or a Regular Expression <code>\d+</code>.</p>
<p>In this case the editor should simply highlight the line indicating a syntax error (it no longer matches any step definitions).</p>
<p>You can read more about the LSP snippet syntax <a href="https://microsoft.github.io/language-server-protocol/specifications/specification-3-17/#snippet_syntax">here</a>.</p>
GitHub
2021-07-13 21:29:09
[cucumber/common] Comment on #1654
<p><a href="https://github.com/davidjgoss">@davidjgoss</a> Our project has multiple jsons from which we send test data to the underlying protractor tests. It was our experience that compile completed faster when we used "tsc --resolveJsonModule false". No other reason as such.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-14 00:32:56
[cucumber/common] Comment on #1653
<p><a href="https://github.com/gasparnagy">@gasparnagy</a> wrote:</p>
<p>> For SpecFlow we would need to also record whether the particular step definition can be used as Given, When, Then or all -- but this can be added later.</p>
<p>I think the best simplest way to handle this is to build three sets of <code>StepDocument</code> - one for each of Given, When and Then.</p>
<p>These three sets of documents can then be added to the four indexes you describe. When the editor sends an auto-complete request, it can send context about what index it wants to search, and the LSP server can use that information to look in the corresponding index.</p>
<p>If the document sets are to be built from Cucumber Messages, we'll have to add a new <code>given|when|then</code> enum field on the <a href="https://github.com/cucumber/common/blob/main/messages/messages.md#stepdefinition">StepDefinition</a> message. The <a href="https://github.com/cucumber/common/blob/main/messages/messages.md#step">Step</a> message already has a <code>keyword</code> field (which would have to be mapped back to the <code>given|when|then</code> enum.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-14 00:33:37
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-14 01:10:24
[cucumber/common] Comment on #1653
<p>Nice!</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-14 02:08:48
[cucumber/common] Comment on #1653
<p>I finally took the time to look at this. It is really cool (I also love the executable README, btw 😉)!</p>
<p>I can't wait to get auto completion in VScode!</p>
GitHub
2021-07-14 20:16:49
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
webpack-contrib/css-loader
<strong><a href="https://togithub.com/webpack-contrib/css-loader/blob/master/CHANGELOG.md#%E2%80%8B600-httpsgithubcomwebpack-contribcss-loadercomparev527v600-2021-07-14|<code">https://togithub.com/webpack-contrib/css-loader/blob/master/CHANGELOG.md#%E2%80%8B600-httpsgithubcomwebpack-contribcss-loadercomparev527v600-2021-07-14|<code</a>v6.0.0</code>></strong></p>
<p><a href="https://togithub.com/webpack-contrib/css-loader/compare/v5.2.7...v6.0.0">Compare Source</a></p>
<p><strong>Notes</strong></p>
<p>• using <code>~</code> is deprecated when the <code>esModules</code> option is enabled (enabled by default) and can be removed from your code (<strong>we recommend it</strong>) (<code>url(~package/image.png)</code> -> <code>url(package/image.png)</code>, <code>@import url(~package/style.css)</code> -> <code>@import url(package/style.css)</code>, <code>composes: import from '~package/one.css';</code> -> <code>composes: import from 'package/one.css';</code>), but we still support it for historical reasons. Why you can removed it? The loader will first try to resolve <code>@import</code>/<code>url()</code>/etc as relative, if it cannot be resolved, the loader will try to resolve <code>@import</code>/<code>url()</code>/etc inside <a href="https://webpack.js.org/configuration/resolve/#resolvemodules"><code>node_modules</code> or modules directories</a>.</p>
<p><strong>⚠️ BREAKING CHANGES</strong></p>
<p>• minimum supported <code>Node.js</code> version is <code>12.13.0</code>
• minimum supported <code>webpack</code> version is <code>5</code>, we recommend to update to the latest version for better performance
• for <code>url</code> and <code>import</code> options <code>Function</code> type was removed in favor <code>Object</code> type with the <code>filter</code> property, i.e. before <code>{ url: () =&gt; true }</code>, now <code>{ url: { filter: () =&gt; true } }</code> and before <code>{ import: () =&gt; true }</code>, now <code>{ import: { filter: () =&gt; true } }</code>
• the <code>modules.compileType</code> option was removed in favor the <code>modules.mode</code> option with <code>icss</code> value, also the <code>modules</code> option can have <code>icss</code> string value
• <code>new URL()</code> syntax used for <code>url()</code>, only when the <code>esModules</code> option is enabled (enabled by default), it means you can bundle CSS for libraries
• <a href="https://developer.mozilla.org/en-US/docs/Web/HTTP/Basics_of_HTTP/Data_URIs">data URI</a> are handling in <code>url()</code>, it means you can register loaders for them, <a href="https://webpack.js.org/configuration/module/#rulescheme">example</a>
• aliases with <code>false</code> value for <code>url()</code> now generate empty data URI (i.e. <code>data:0,</code>), only when the <code>esModules</code> option is enabled (enabled by default)
• <code>[ext]</code> placeholder don't need <code>.</code> (dot) before for the <code>localIdentName</code> option, i.e. please change <code>.[ext]</code> on <code>[ext]</code> (no dot before)
• <code>[folder]</code> placeholder was removed without replacement for the <code>localIdentName</code> option, please use a custom function if you need complex logic
• <code>[emoji]</code> placeholder was removed without replacement for the <code>localIdentName</code> option, please use a custom function if you need complex logic
• the <code>localIdentHashPrefix</code> was removed in favor the <code>localIdentHashSalt</code> option</p>
<p><strong>Features</strong></p>
<p>• supported <a href="https://webpack.js.org/configuration/resolve/#resolvebydependency"><code>resolve.byDependency.css</code></a> resolve options for <code>@import</code>
• supported <a href="https://webpack.js.org/configuration/resolve/#resolvebydependency"><code>resolve.byDependency.icss</code></a> resolve CSS modules and ICSS imports (i.e. <code>composes</code>/etc)
• added <code>modules.localIdentHashFunction</code>, <code>modules.localIdentHashDigest</code>, <code>modules.localIdentHashDigestLength</code> options for better class hashing controlling
• less dependencies</p>
<p><strong>Bug Fixes</strong></p>
<p>• better performance
• fixed circular <code>@import</code></p>
<p><strong>Notes</strong></p>
<p>• <strong>we strongly recommend not to add <code>.css</code> to <code>resolve.extensions</code>, it reduces performance and in most cases it is simply not necessary, alternative you can set resolve options <a href="https://webpack.js.org/configuration/resolve/#resolvebydependency">by dependency</a></strong></p>
<p><strong><a href="https://togithub.com/webpack-contrib/css-loader/compare/v5.2.6...v5.2.7">5.2.7</a> (2021-07-13)</strong>
<strong>Bug Fixes</strong></p>
<p>• fix crash when source map is unavailable with external URL in <code>[@import](<https://togithub.com/import>)</code> (<a href="https://togithub.com/webpack-contrib/css-loader/commit/bb76fe48a198e74cacf29ad4b1c01d485f4db11f">bb76fe4</a>)</p>
<p><strong><a href="https://togithub.com/webpack-contrib/css-loader/compare/v5.2.5...v5.2.6">5.2.6</a> (2021-05-24)</strong>
<strong>Bug Fixes</strong></p>
<p>• always write locals export when css modules/icss enabled (<a href="https://togithub.com/webpack-contrib/css-loader/issues/1315">#1315</a>) (<a href="https://togithub.com/webpack-contrib/css-loader/commit/075d9bd044a78543479cbf10ccd3c386a3e434e6">075d9bd</a>)</p>
<p><strong><a href="https://togithub.com/webpack-contrib/css-loader/compare/v5.2.4...v5.2.5">5.2.5</a> (2021-05-20)</strong>
<strong>Bug Fixes</strong></p>
<p>• compatibility with named export and es5 (<a href="https://togithub.com/webpack-contrib/css-loader/issues/1314">#1314</a>) (<a href="https://togithub.com/webpack-contrib/css-loader/commit/0cf8cdedd8667b1ba13d3b4322087943a25176f6">0cf8cde</a>)</p>
<p><strong><a href="https://togithub.com/webpack-contrib/css-loader/compare/v5.2.3...v5.2.4">5.2.4</a> (2021-04-19)</strong>
<strong>Bug Fixes</strong></p>
<p>• do not crash on 'false' aliases (<a href="https://togithub.com/webpack-contrib/css-loader/issues/1292">#1292</a>) (<a href="https://togithub.com/webpack-contrib/css-loader/commit/e913cb1d73a4f5c3c4464e0446a885e9f677a005">e913cb1</a>)</p>
<p><strong><a href="https://togithub.com/webpack-contrib/css-loader/compare/v5.2.2...v5.2.3">5.2.3</a> (2021-04-19)</strong>
<strong>Bug Fixes</strong></p>
<p>• improve performance</p>
<p><strong><a href="https://togithub.com/webpack-contrib/css-loader/compare/v5.2.1...v5.2.2">5.2.2</a> (2021-04-16)</strong>
<strong>Bug Fixes</strong></p>
<p>• avoid escape nonASCII characters in local names (<a href="https://togithub.com/webpack-contrib/css-loader/commit/072273308a8ab4b7efdae31440689dc81978ca1d">0722733</a>)</p>
<p><strong><a href="https://togithub.com/webpack-contrib/css-loader/compare/v5.2.0...v5.2.1">5.2.1</a> (2021-04-09)</strong>
<strong>Bug Fixes</strong></p>
<p>• do not crash on unescaped svg data uri (<a href="https://togithub.com/webpack-contrib/css-loader/issues/1288">#1288</a>) (<a href="https://togithub.com/webpack-contrib/css-loader/commit/4f289c5e4df6c666fdf6dd3402560ae74d4bf7ee">4f289c5</a>)</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
GitHub
2021-07-15 03:10:33
[cucumber/common] Pull request opened by aslakhellesoy
<p><strong>Summary</strong></p>
<p>Change the <code>StepDocument</code> so it is an object with a <code>suggestion</code> string and a <code>segments</code> array.</p>
<p><strong>Motivation and Context</strong></p>
<p>We don't want to display the actual LSP snippet in the suggestions the user sees. The LSP snippet should never be<br />
seen by users - it's only passed to the editor to build the inserted text.</p>
<p>By adding a <code>suggestion</code> string, we can display something nicer to the user, such as <code>I have {int} cukes in my {word}</code>.</p>
<p><strong>Types of changes</strong></p>
<p>☐ Bug fix (non-breaking change which fixes an issue).
☐ New feature (non-breaking change which adds functionality).
☑︎ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<p><strong>Checklist:</strong></p>
<p>☐ The change has been ported to Java.
☐ The change has been ported to Ruby.
☑︎ The change has been ported to JavaScript.
☐ The change has been ported to Go.
☐ The change has been ported to .NET.
☐ I've added tests for my code.
☐ My change requires a change to the documentation.
☐ I have updated the documentation accordingly.
☐ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-15 03:44:53
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-15 03:56:21
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-15 06:23:55
[cucumber/common] Pull request opened by aslakhellesoy
<p><strong>Summary</strong></p>
<p>Add custom rendering support for <code>&lt;Feature&gt;</code></p>
<p><strong>Motivation and Context</strong></p>
<p>I want to render a button/link on top of every feature that when clicked, displays an editor.</p>
<p>I exposed the <code>Feature.DefaultRenderer</code> so that the overriding component can still use the original. Is this something that we should do for the other components too?</p>
<p><strong>How Has This Been Tested?</strong></p>
<p>Using storybook</p>
<p><strong>Screenshots (if appropriate):</strong>
<strong>Types of changes</strong></p>
<p>☐ Bug fix (non-breaking change which fixes an issue).
☑︎ New feature (non-breaking change which adds functionality).
☐ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<p><strong>Checklist:</strong></p>
<p>☐ The change has been ported to Java.
☐ The change has been ported to Ruby.
☑︎ The change has been ported to JavaScript.
☐ The change has been ported to Go.
☐ The change has been ported to .NET.
☐ I've added tests for my code.
☐ My change requires a change to the documentation.
☐ I have updated the documentation accordingly.
☐ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-15 06:25:35
[cucumber/common] Pull request opened by davidjgoss
<p><strong>Summary</strong></p>
<p>Add cases to the compatibility kit for:</p>
<p>• Pending steps
• Skipped scenarios via hooks/steps
• Undefined steps</p>
<p><strong>Motivation and Context</strong></p>
<p>We want to document and standardise the behaviour of these features. We'll also need them in order to use the CCK as the reference point for a JUnit formatter (see <a href="https://github.com/cucumber/common/issues/871">#871</a>).</p>
<p><strong>How Has This Been Tested?</strong></p>
<p>Test added to fake-cucumber which now supports <code>skipped</code>.</p>
<p><strong>Types of changes</strong></p>
<p>☐ Bug fix (non-breaking change which fixes an issue).
☑︎ New feature (non-breaking change which adds functionality).
☐ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<p><strong>Checklist:</strong></p>
<p>☐ The change has been ported to Java.
☑︎ The change has been ported to Ruby.
☐ The change has been ported to JavaScript.
☐ The change has been ported to Go.
☐ The change has been ported to .NET.
☑︎ I've added tests for my code.
☐ My change requires a change to the documentation.
☐ I have updated the documentation accordingly.
☑︎ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-15 07:20:26
[cucumber/common] Issue reopened by aslakhellesoy
<https://github.com/cucumber/common/issues/590|#590 Gherkin: Remove Background from Rule in grammar>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-15 07:21:28
[cucumber/common] Comment on #590
<p>Reopening this because:</p>
<p>• We still think this is a bad idea
• The SpecFlow team doesn't want to implement it</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-15 11:23:39
[cucumber/common] Comment on #1659
<p>Dropping the two regular steps would simply this.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-15 11:25:27
[cucumber/common] Comment on #1659
<p>Same as before. There are a few more.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-15 11:29:35
[cucumber/common] Comment on #1659
<p>I wanted the messages that come out of this to show that the step before runs and passes, and the step after is skipped - I think that's key to properly implementing this?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-15 11:34:04
[cucumber/common] Comment on #1659
<p>Maybe I should make this explicit in the wording of these steps?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-15 12:39:31
[cucumber/common] Comment on #1659
<p>> I wanted the messages that come out of this to show that the step before runs and passes, and the step after is skipped - I think that's key to properly implementing this?</p>
<p>I think that should be a separate test case then. Could be much more generic. Any step after a non-passing step is skipped.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-15 16:17:17
[cucumber/common] Pull request opened by aslakhellesoy
<p><strong>Summary</strong></p>
<p>Ensure all components in <code>@cucumber/react</code></p>
<p><strong>Details</strong></p>
<p>This changes most components to <code>export const SomeComponent = ...</code> and removes the <code>export default SomeComponent</code>.<br />
The motivation for this is to more easily export everything in a directory, and do it in a nested fashion.</p>
<p>From the consumer's perspective, this means components are now imported like this:</p>
<pre><code>import { components } from '@cucumber/react'
const { DocString, Tags } = components.gherkin
</code></pre>
<p>This is a bit more verbose than the previous:</p>
<pre><code>import { DocString, Tags } from '@cucumber/react'
</code></pre>
<p>But since this library has grown quite big, I feel it's a bit tidier to have a bit more namespacing rather than cramming everything into the top-level import. Trying to do that also makes it easier to forget to export something, whereas now, the various directories' <code>index.ts</code> spreads this responsibility over multiple files, so it's easier to spot an omission.</p>
<p>I've also made the <code>&lt;Feature&gt;</code> component styleable.</p>
<p><strong>TODO before merging</strong></p>
<p>☐ Make all components customisable with component overrides
☐ Move all styling into scss
☐ Demonstrate how to override component styles with TailwindCSS by defining classes with <code>@apply</code> in a Storybook story
☐ Get rid of all usages of <code>components.gherkin.SomeComponent</code> and use destructuring at the top of the file
☐ Change <code>filter</code> so we can do <code>import { filter } from '@cucumber/react'</code>
☐ Move some files around to avoid up-and-down dependencies between directories</p>
<p><strong>Motivation and Context</strong></p>
<p>Make sure everything is exported and make things a bit more organised</p>
<p><strong>How Has This Been Tested?</strong></p>
<p>With StoryBook and tsc - and existing tests are still passing.</p>
<p><strong>Types of changes</strong></p>
<p>☐ Bug fix (non-breaking change which fixes an issue).
☐ New feature (non-breaking change which adds functionality).
☑︎ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<p><strong>Checklist:</strong></p>
<p>☐ The change has been ported to Java.
☐ The change has been ported to Ruby.
☑︎ The change has been ported to JavaScript.
☐ The change has been ported to Go.
☐ The change has been ported to .NET.
☐ I've added tests for my code.
☐ My change requires a change to the documentation.
☐ I have updated the documentation accordingly.
☐ I have updated the CHANGELOG accordingly.</p>
GitHub
2021-07-15 16:22:40
[cucumber/common] Comment on #1658
<p>Superseded by <a href="https://github.com/cucumber/common/pull/1660">#1660</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-15 16:22:41
[cucumber/common] Pull request closed by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-16 00:59:23
[cucumber/common] Comment on #1659
<p>I kind of prefer to have the test cases for pending etc be fairly complete even if there is a little duplication overall with the skip-subsequent-steps concept, especially as there is some patchiness in our implementations (e.g. ruby doesn't support pending or skipped it seems).</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-16 01:12:24
[cucumber/common] Comment on #1660
<p>I think this may be used in more places than the <code>&lt;Feature&gt;</code> component?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-16 01:12:24
[cucumber/common] davidjgoss commented on aslakhellesoy's pull request
<p>Looks good, agree with the approach. We could also look at enabling something like:</p>
<pre><code>import { DocString } from '@cucumber/react/components/gherkin'
</code></pre>
<p>I think that would require us being very deliberate about the file structure of the compiled output and <a href="https://api-extractor.com/pages/setup/configure_rollup/">the <code>.d.ts</code> files</a> but would probably be possible. It might come down to which (if any) is more reliable inferred by editors?</p>
<p>> Make all components customisable with component overrides</p>
<p>Do we really mean every component in here? Some are a bit in flux (especially the "chrome" stuff that we have yet to restyle/redesign, plus the accordion might be replaced as you've said before) and I'd be wary of effectively expanding the public API surface more than necessary if there's not an obvious use case.</p>
<p>> Demonstrate how to override component styles with TailwindCSS by defining classes with <a href="https://github.com/apply">@apply</a> in a Storybook story</p>
<p>My next task on <a href="https://github.com/cucumber/common/issues/1592">#1592</a> is to add tailwind to the tooling - I can do a standalone PR for that unless you have already started?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-16 02:44:56
[cucumber/common] Comment on #1659
<p>I understand, but when the actual and expected result differ it is very hard to to see what the problem is when there is extra fluff laying around.</p>
<p>Consider instead using multiple examples:</p>
<pre><code> Scenario: Unimplemented step signals pending status
Given a step that isnt implemented yet
Scenario: Steps after unimplemented steps are skipped
Given a step that isnt implemented yet
Then a skipped step
Scenario: Steps before unimplemented steps are executed
Given an implemented step
Then a step that isnt implemented yet
</code></pre>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-16 09:12:11
[cucumber/common] Comment on #1660
<p>Fixed in <a href="https://github.com/cucumber/common/commit/53cc7f6b27720d4d9aa9fff14cde50bf7097f306">53cc7f6</a> - added a <code>&lt;Children&gt;</code> component</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-16 09:12:45
[cucumber/common] Comment on #1660
<p>> My next task on <a href="https://github.com/cucumber/common/issues/1592">#1592</a> is to add tailwind to the tooling</p>
<p>Ah yes, removed the tailwind TODO from this PR</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-16 10:17:38
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
webpack/webpack
<strong><a href="https://togithub.com/webpack/webpack/releases/v5.45.0"><code>v5.45.0</code></a></strong></p>
<p><a href="https://togithub.com/webpack/webpack/compare/v5.44.0...v5.45.0">Compare Source</a></p>
<p><strong>Features</strong></p>
<p>• add support to import assertions</p>
<p><strong>Bugfixes</strong></p>
<p>• SourceMaps will now also be added to <code>.cjs</code> output files
• fix non-system externals in a system library</p>
<p><strong>Performance</strong></p>
<p>• avoid copying timestamps from the watcher to the compiler</p>
<p><strong>Contributing</strong></p>
<p>• update to jest 27</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled due to failing status checks.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-16 10:17:39
[cucumber/common] Comment on #1661
<p><strong>Branch automerge failure</strong></p>
<p>This PR was configured for branch automerge, however this is not possible so it has been raised as a PR instead.</p>
<hr />
<p>• Branch has one or more failed status checks</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-16 11:31:09
[cucumber/common] Issue opened by nddipiazza
<p>I find myself wishing there was a simple SDK with the ability to pretty print (format) gherkin files.</p>
<p>Like how intellij cucumber plugin formats them is exactly what I want.</p>
<p>Is there some way to add a simple static method for this?</p>
<p><code>GherkinFormatter.prettyPrint(gherkinString)</code></p>
<p>I think there is a way to do this with the library, but it's not as simple.</p>
GitHub
2021-07-16 11:32:23
[cucumber/common] Comment on #1662
<p>I've moved this to the common repo where gherkin is maintained.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-16 15:14:44
[cucumber/common] Pull request merged by renovate[bot]
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-16 23:54:31
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
Qix-/color
<strong><a href="https://togithub.com/Qix-/color/compare/3.2.0...4.0.0"><code>v4.0.0</code></a></strong></p>
<p><a href="https://togithub.com/Qix-/color/compare/3.2.0...4.0.0">Compare Source</a></p>
<p><strong><a href="https://togithub.com/Qix-/color/compare/3.1.4...3.2.0"><code>v3.2.0</code></a></strong></p>
<p><a href="https://togithub.com/Qix-/color/compare/3.1.4...3.2.0">Compare Source</a></p>
<p><strong><a href="https://togithub.com/Qix-/color/releases/3.1.4"><code>v3.1.4</code></a></strong></p>
<p><a href="https://togithub.com/Qix-/color/compare/3.1.3...3.1.4">Compare Source</a></p>
<p><strong>Patch Release 3.1.4</strong></p>
<p>• Bumped <code>color-string</code> to latest version, resolving problems for users using <code>package-lock.json</code> files and adding support for space-separated <code>hsl()</code> syntax.</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-17 09:15:13
[cucumber/common] Comment on #590
<p><em>continues to swing on his excess rope enabled tire swing while pondering the implications of having all language platforms' behavior being dictated by the lowest common denominator of language platform aggreement</em></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-17 09:25:53
[cucumber/common] Comment on #590
<p>> I had a crack at this (removed <code>Background?</code> from the <code>Rule</code> grammar rule).
>
> Because of the way Berp works, a rule with a background does <em>not</em> cause a parser error. It just generates a <code>Rule</code> AST node with description <code>Background:\n Given rb</code>.
>
> The Berp parser will only attempt to match lines it expects. Everything else becomes a description. I think parser is unable to detect quite a lot of parse errors. /cc <a href="https://github.com/gasparnagy">@gasparnagy</a></p>
<p>I think the easiest way to implement this is to allow Background after Rule in the parser, but throw an exception in the AST builder.</p>
GitHub
2021-07-17 09:27:25
[cucumber/common] Comment on #590
<p>Hmmm, I find it quite useful, and use it fairly regularly - if background makes sense for a feature as a whole, why could it not make sense at a rule level, when a rule has several examples associated with it?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-17 10:37:36
[cucumber/common] Comment on #590
<p><a href="https://github.com/wakaleo">@wakaleo</a> There's an anti-Background sentiment in some quarters anyhow (see Seb and Gaspar's last book) so extending it to Rules could be seen as making things worse</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-17 10:54:19
[cucumber/common] Comment on #590
<p>Hmm. it's my understanding that versions of Cucumber have been released with the <code>Rule</code> keyword supporting <code>Background</code>, right? If that's the case, then isn't removing <code>Background</code> from <code>Rule</code> just as invasive as removing <code>Background</code> altogether?</p>
<p>I mean to ask: what alternative do we offer those who already invested in either <code>Background</code> heavily? Either because they shot themselves in the foot or because they genuinely had a good reason to use the Background abstraction?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-17 11:11:14
[cucumber/common] Comment on #590
<p>> Hmm. it's my understanding that versions of Cucumber have been released with the <code>Rule</code> keyword supporting <code>Background</code>, right? If that's the case, then isn't removing <code>Background</code> from <code>Rule</code> just as invasive as removing <code>Background</code> altogether?</p>
<p>Ok. So, I just figured out that the <code>Rule</code> keyword got the <code>Background</code> since its inception (Gherkin v6.0). Gherkin 6 was released in 2018 (or at least the merge did). I'd argue that it's simply too late to remove the background from the Rule keyword. <a href="https://github.com/gasparnagy">@gasparnagy</a> provides the perfect reason in his original comment explaining why he thinks it shouldn't have been added in the first place:</p>
<p>> with OSS projects that are used by so many people.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-17 13:51:10
[cucumber/common] Comment on #590
<p>We have had backgrounds in rules for a while now.</p>
<p>While we speak of being opinionated, we, its creators clearly thought it was not important to fix the parser. In fact, we even removed the experimental warning from the documentation! If we are opinionated, we are also wishy washy in what our opinion is!</p>
<p>And that is a problem. And we should probably have a discussion about how we got here in the first place.</p>
<p>But form a practical point of view:</p>
<ol><li>Removing background rules will have significant user impact.</li><li>Backgrounds rules are hidden from the rest of Cucumber through pickles. So the complexity of keeping background rules is contained.</li><li>Removing background rules will require a deprecation process that spans two major versions of each Cucumber implementation. One to deprecate, one to remove. This will take a significant amount of time and work to coordinate.</li>
</ol>
<p>So if given the choice between keeping things as is, breaking background rules abruptly or deprecating them I'd say that leaving things as is is the least amount of effort and doesn't annoy users.</p>
<p>> The SpecFlow team doesn't want to implement it</p>
<p><a href="https://github.com/gasparnagy">@gasparnagy</a> could you elaborate on the why here. We haven't really discussed the reasons and motivations here.</p>
GitHub
2021-07-17 14:18:23
[cucumber/common] Comment on #1386
<p>The current Perl implementation doesn't have a runtime dependency on protobuf. Does that mean I can remove the <code>language:perl</code> tag, or is there additional work that this issue has turned into now?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-17 14:23:28
[cucumber/common] Comment on #768
<p>I'll be joining the stand-up coming Thursday. I'm keen to understand what will be required to bring this forward, because it's one of the last obstacles for <code>Test::BDD::Cucumber</code> to move to using the Gherkin library.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-17 14:56:45
[cucumber/common] Comment on #590
<p>There are many good arguments made here for keeping it. Another way to address this is through documentation - make it loud and clear that using backgrounds in rules is <em>supported</em>, but potentially a bad idea. I think removing it would be akin to adding a limit of max 1000 lines for a source file - a bit patronising.</p>
<p><a href="https://i.imgur.com/E1XYHtc.gifv">You can't shut the stable door after the horse has bolted</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-17 15:04:14
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-17 15:39:08
[cucumber/common] Comment on #768
<p>In response to <a href="https://github.com/ciaranmcnulty">@ciaranmcnulty</a> :</p>
<p>> If this goes ahead we'll need to consider / test the impact on i18n
>
> Will it add to my experience to have to remember to use <code>Given()</code> whenever I write <code>Gegewe</code> in my steps?</p>
<p>That depends: in the Perl Cucumber variant that I maintain, translations for Given/When/Then are registered for the step files to be available (documentation: <a href="https://metacpan.org/pod/Test::BDD::Cucumber::Manual::Tutorial#Non-English-features-and-step-files">https://metacpan.org/pod/Test::BDD::Cucumber::Manual::Tutorial#Non-English-features-and-step-files</a>). No idea how other implementations would (be able to) deal with this though.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-17 16:02:25
[cucumber/common] Issue opened by aslakhellesoy
<p><strong>Is your feature request related to a problem? Please describe.</strong></p>
<p>Exposing the right API from TypeScript modules is tricky as we've discussed in <a href="https://github.com/cucumber/common/pull/1660#pullrequestreview-708109485">#1660 (review)</a></p>
<p><strong>Describe the solution you'd like</strong></p>
<p>Adopt <a href="https://api-extractor.com">API Extractor</a> which solves this problem</p>
<p><strong>Describe alternatives you've considered</strong></p>
<p>Using a homegrown namespace as currently implemented in <a href="https://github.com/cucumber/common/pull/1660">#1660</a>, but that doesn't really solve the "we forgot to export this file" problem.</p>
<p><strong>Additional context</strong></p>
<p>This is also related to <a href="https://github.com/mpkorstanje">@mpkorstanje</a>'s recent soapbox talks about having stable public APIs.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-18 03:03:57
[cucumber/common] Pull request opened by ehuelsmann
<p><strong>Summary</strong></p>
<p>Miscelaneous cleanup made possible after upgrading the Perl minimum requirement to 5.12.0 (which was released in April 2010).</p>
<p><strong>Details</strong></p>
<p>• Since 5.12, Perl can open filehandles to in-memory byte sequences. Remove an external depenency (<code>IO::Scalar</code>) by taking advantage of that.
• Additionally, since 5.12, all file handles are blessed references into <code>IO::File</code>, which removes the need to explicitly load and create <code>IO::File</code> instances.
• Close the input file as soon as the last line is read, instead of when the TokenScanner goes out of scope; it's simply better form to do that early</p>
<p><strong>Motivation and Context</strong></p>
<p>The resulting code is overall more clean.</p>
<p>This change, although it runs correctly over the test data set, <em>may</em> fix a bug with empty lines being considered the end of the input file. (an empty string is considered "false" and when <code>$line</code> evaluated to false, an "undef" result was returned, signalling end-of-file...; but this makes me wonder why the test suite succeeded.)</p>
<p><strong>How Has This Been Tested?</strong></p>
<p>The test suite runs on this change as expected.</p>
<p><strong>Types of changes</strong></p>
<p>☑︎ Bug fix (non-breaking change which fixes an issue).
☑︎ Dependency update.</p>
<p><strong>Checklist:</strong></p>
<p>☑︎ The change does not need to be ported to other implementations
☑︎ Existing tests cover this change.
☑︎ My change does not require a change to the documentation.
☑︎ No documentation changes are required.
☐ I have updated the CHANGELOG accordingly.</p>
GitHub
2021-07-18 06:52:33
[cucumber/common] Comment on #1646
<p>The Perl step is one of the steps that take longest to run. I've submitted changes to the Dockerfile which could reduce Perl build time. I'm not sure where <em>inside</em> the Perl job the time is spent exactly (and why that would take this amount of time when the e.g. Ruby one does not). However, no amount of reducing Perl time will significantly reduce the 50 minutes. There are just a lot of components here...</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-19 01:42:24
[cucumber/common] Comment on #1664
<p>This will be great! I did try to add it a while ago and ran into some issues, but IIRC it was because of some peculiarities in the TypeScript code emitted by protobufjs, so hopefully smooth sailing this time.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-19 02:03:58
[cucumber/common] Pull request merged by aurelien-reeves
<https://github.com/cucumber/common/pull/1655|#1655 Convert ruby messages duration.nanos to integer>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-19 05:55:04
[cucumber/common] Comment on #1662
<p>There is a <code>pretty</code> method implemented in javascript. Would that fit your needs?</p>
<p>There is also <a href="https://github.com/cucumber/common/pull/1555">#1555</a> which work in enhancing that <code>pretty</code> method.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-19 05:56:53
[cucumber/common] Comment on #1386
<p>> The current Perl implementation doesn't have a runtime dependency on protobuf. Does that mean I can remove the <code>language:perl</code> tag, or is there additional work that this issue has turned into now?</p>
<p>Thanks for the tip. I've removed the label.</p>
<p>Also, we may audi the ruby code, but I think it is now free from any dependency to protobuf.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-19 08:35:46
[cucumber/common] Pull request opened by aurelien-reeves
<p>Go libraries without executable was actually not properly released</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-19 08:44:59
[cucumber/common] New release Release - gherkin/go/v20.0.1 published by cukebot
GitHub
2021-07-19 09:04:21
[cucumber/common] Comment on #590
<p>In terms of the backwards compatibility worries, I know the parser allows them but has there been any user-facing documentation mentioning/encouraging Backgrounds in Rules? If it was never documented as working then arguably it's a bug fix :)</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-19 09:06:24
[cucumber/common] Comment on #590
<p><a href="https://github.com/ciaranmcnulty">@ciaranmcnulty</a> whoops <a href="https://github.com/cucumber/docs/pull/440">cucumber/docs#440</a></p>
<p>> make it loud and clear that using backgrounds in rules is supported, but potentially a bad idea.</p>
<p>I think the documentation added in the PR above makes the statement but doesn't really get to the <em>why</em> it's a bad idea and what you should do instead.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-19 09:06:46
[cucumber/common] Comment on #590
<p>> In terms of the backwards compatibility worries, I know the parser allows them but has there been any user-facing documentation mentioning/encouraging Backgrounds in Rules? If it was never documented as working then arguably it's a bug fix :)</p>
<p>I believe I mention it in BDD in Action (2nd Edition) if that counts as user-facing documentation.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-19 09:22:00
[cucumber/common] Comment on #590
<p>> > The SpecFlow team doesn't want to implement it
>
> <a href="https://github.com/gasparnagy">@gasparnagy</a> could you elaborate on the why here. We haven't really discussed the reasons and motivations here.</p>
<p>This is nothing that <a href="https://github.com/gasparnagy">@gasparnagy</a> said <a href="https://github.com/mpkorstanje">@mpkorstanje</a>. That was me and the rest of the SpecFlow team at Tricentis.<br />
Our motivation was the following: The support for Rules isn't complete in SpecFlow and all our tools. We want to tackle this finally. And so the topic of Background in Rules popped up for us because it is one thing that we don't support yet.<br />
For me the decision we did 2019 (or was is 2018) that we will remove the Background of Rules is still the last decision about this topic and was validated last Thursday again.<br />
So in our case, we also need now to do less to support Rules completely.<br />
That was our motivation. Only do the work that is really needed.</p>
<p>But if we say, it is now too late to remove the Background. This is also fine for us. Then we will implement it in SpecFlow. We just didn't want to implement something, when we said we will remove it, and is afaik not been implemented in every implementation yet.</p>
<p>We just need now a final decision on what we do with the Background in Rules.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-19 10:08:09
[cucumber/common] Comment on #1662
<p>I had originally opened this ticket in the cucumber-jvm project. it's there that I really need this. this was moved to the generic cucumber project which loses that fact.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-19 10:50:59
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about these updates again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-19 23:55:14
[cucumber/common] Comment on #1662
<p>I considered the fact that you opened it in cucumber-jvm first.<br />
Maybe the javascript version for gherkin-utils would had fit your needs</p>
GitHub
2021-07-19 23:57:43
[cucumber/common] Comment on #1667
<p>The dependency to cucumber 4.1 in <code>json-to-messages/ruby-testdata</code>: is it still a requirement?<br />
All tests seems to pass with cucumber 7.0</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 01:31:57
[cucumber/common] Comment on #1667
<p>This was something to do with the fact that format json was being deprecated, but now we're basically maintaining it I think this is fine.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 06:19:14
[cucumber/common] Comment on #1662
<p>Did you consider the possibility to use cucumber-jvm to get the messages in the form of ndjson to use those as an input in javascript implementation of gherkin and then get the gherkinDocument to use with gherkin-utils?</p>
GitHub
2021-07-20 06:43:03
[cucumber/common] Comment on #1653
<p>Finally had some time to take a look</p>
<p>Indeed, the executable readme, this is really great!</p>
<p>Regarding <code>suggest</code> itself, I am concerned about the memory consumption for StepDocuments.</p>
<p>I may say something silly: maybe we could bring sqlite as storage engine? We could even benefit from its full-text search engine! Would that be doable?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 07:27:19
[cucumber/common] Comment on #1653
<p>At what memory size would you consider consumption to be problematic <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a>?</p>
<p>Roughly how many (by order of magnitude) <code>StepDocument</code> objects would you estimate would be needed to reach that critical level?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 07:31:50
[cucumber/common] Issue opened by customcommander
<p><em>I'm sure it must have been suggested before but I cannot find a similar ticket in this repo. Apologies in advance if this discussion has already been settled.</em></p>
<hr />
<p><strong>Is your feature request related to a problem? Please describe.</strong></p>
<p>We have naming convention for tags e.g.</p>
<pre><code>@featureAAA
@featureBBB
@featureCCC
@regressionAAA
@regressionBBB
@regressionCCC
</code></pre>
<p>We apply these tags on scenarios across many files. We can't have all the features scenarios in one directory and all the regressions scenarios in another. We need to be able to mix and match.</p>
<p>With such a setup the only way to run either all features or all regressions scenarios is to apply a general tag <code>@feature</code> and <code>@scenario</code> as well.</p>
<p><strong>Describe the solution you'd like</strong></p>
<p>Would we nice if we could have a tag expression such as <code>@^feature</code> which means "all tags that starts with <code>feature</code>"</p>
<p><strong>Describe alternatives you've considered</strong></p>
<p>The alternative is the apply a global tag as well e.g.</p>
<pre><code>@featureAAA @feature
Scenario: …
Given …
When …
Then …
</code></pre>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 07:39:06
[cucumber/common] Comment on #1668
<p>May I ask, why applying several tags instead of one composed tag does not fit your needs?</p>
<p>In your examples, I would apply the following tags:</p>
<pre><code>@feature @AAA
@feature @BBB
@feature @CCC
@regression @AAA
@regression @BBB
@regression @CCC
</code></pre>
<p>That way you could even have more possibilities when properly using <a href="https://github.com/cucumber/common/tree/main/tag-expressions">tag-expressions</a> like things like <code>@feature or @AAA</code>, or <code>@AAA or (@BBB and not @regression)</code></p>
<p>You are actually not limited into the number of tags you can have on features, rules, or examples</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 07:42:29
[cucumber/common] Comment on #1653
<p>We would have to benchmark to have proper idea actually</p>
<p>That was just a rough idea</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 07:45:19
[cucumber/common] Comment on #1668
<p>Hey thanks for the quick reply.</p>
<p>Sorry my example above should have been: (so each tag is unique)</p>
<pre><code>@featureAAA
@featureBBB
@featureCCC
@regressionDDD
@regressionEEE
@regressionFFF
</code></pre>
<p>(We need unique tags because of integration with other tools.)</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 07:47:03
[cucumber/common] Comment on #1668
<p>> (We need unique tags because of integration with other tools.)</p>
<p>Ok, thanks for the precision</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 07:54:47
[cucumber/common] Comment on #1653
<p>> At what memory size would you consider consumption to be problematic <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a>?</p>
<p>That's a difficult question actually.<br />
If we consider only <code>suggest</code>, we may consider that we could easily take benefit from GB of user-memory and that it won't be an issue.</p>
<p>But actually it will be part of bigger tools like IDEs, or in-browser code editors. Those tools are already huge memory consumer. So I think we may try to reduce that impact on user-memory if this is possible and not too difficult.</p>
<p>As this is just the beginning of the project, that would be a good opportunity to think about that right now as it may be easier than with a more mature project.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 08:13:36
[cucumber/common] Comment on #1662
<p><a href="https://github.com/nddipiazza">@nddipiazza</a> what are you trying to achieve. You've described a solution, but not your problem. What are you trying to achieve by pretty formatting gherkin files?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 08:13:44
[cucumber/common] Comment on #590
<p>I agree with <a href="https://github.com/mpkorstanje">@mpkorstanje</a></p>
<p>I think that if this is there since the beginning of rules, and it has been at some point documented - even if asked to avoid it - we can not assume that user do not have background for their rules.</p>
<p>And if we remove it, we will certainly have plenty of requests to get it back.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 08:18:08
[cucumber/common] Comment on #1668
<p>So you can't add a <code>@feature</code> tag as well <a href="https://github.com/customcommander">@customcommander</a>?</p>
<p>e.g.</p>
<pre><code>@feature
@featureAAA
@featureBBB
@featureCCC
@regression
@regressionDDD
@regressionEEE
@regressionFFF
</code></pre>
<p>I can see how what you're suggesting is useful, but it's always worth pushing back a little to see if we can work with what we have before we add more complexity to the code.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 08:19:03
[cucumber/common] Comment on #590
<p>To put it into perspective, here is an example where multiple backgrounds and rules are useful (Admittedly only one use case).</p>
<p>I also have examples with Backgrounds at the top level and then rule/background. I think since 2018 when we were debating it and 2019 when we accepted it, I've started to find more and more use-cases where it's useful. For me it needs to stay.</p>
<p>If some languages don't want to implement it, I say that's ok. But iirc what <a href="https://github.com/SabotageAndi">@SabotageAndi</a> is saying is true. In 2018 rules were quite new and we were struggling with implementing them. Now they're common place let's implement them wholly >??</p>
<pre><code>Feature: Advice feedback
As a ...
I want to ....
So that I ....
Rule: Rule1
Background:
Given I am logged in
And I am on ....
Scenario: User can give feedback
Then I am able to provide feedback
Scenario: User giving positive feedback
When I inform ...
Then I am thanked for providing feedback
Scenario: User giving detailed free-text negative feedback
When I inform ....
And I offer a typed reason for why the advice was not useful
Then I am thanked for providing feedback
Rule: Rule2
Background:
Given I am on ....
Scenario: User can give feedback
Then I am able to provide feedback
Scenario: User giving positive feedback
When I inform ...
Then I am thanked for providing feedback
Scenario: User giving detailed free-text negative feedback
When I inform ....
And I offer a typed reason for why the advice was not useful
Then I am thanked for providing feedback
Scenario: User giving detailed negative feedback with a specific reason
When I inform ...
And I select a reason that the advice was not useful
Then I am thanked for providing feedback
Scenario: User changing their mind about feedback
When I inform ....
And I change my mind about giving feedback
Then I am able to provide feedback
</code></pre>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 08:37:37
[cucumber/common] Comment on #1668
<p>Hey <a href="https://github.com/mattwynne">@mattwynne</a> absolutely right I understand and I realise that my initial post is not great actually.</p>
<p>So IRL it would look something like this:</p>
<p><em>dir1/file4.feature</em></p>
<pre><code>@featureAAA
Scenario: …
Given …
</code></pre>
<p><em>dir2/file5.feature</em></p>
<pre><code>@featureBBB @regressionDDD
Scenario: …
Given …
</code></pre>
<p><em>dir3/file6.feature</em></p>
<pre><code>@regressionEEE
Scenario: …
Given …
</code></pre>
<p>Notes:</p>
<p>• Each directory may have one or more feature files
• Each feature file may have one or more scenarios
• In any feature file you can have a mixed bag of tags
• Any scenario can be tagged multiple times with any kind of tags</p>
<p>We absolutely could do this:</p>
<p><em>dir1/file4.feature</em></p>
<pre><code>@feature
@featureAAA
Scenario: …
Given …
</code></pre>
<p><em>dir2/file5.feature</em></p>
<pre><code>@feature
@featureBBB
@regression
@regressionDDD
Scenario: …
Given …
</code></pre>
<p><em>dir3/file6.feature</em></p>
<pre><code>@regression
@regressionEEE
Scenario: …
Given …
</code></pre>
<p>But we would like to avoid ;)</p>
<p>In reality the set of files, scenarios and tags is much much larger so relying on engineers to "do the right thing" is risky :-)</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 09:02:39
[cucumber/common] Comment on #1668
<p><a href="https://github.com/customcommander">@customcommander</a></p>
<p>So if I understand it correctly for your developers you currently have the following rules:</p>
<p>• Each scenario must be tagged.
• The tag must be of the format <code>@feature[A-Z]**</code>, <code>@regression[A-Z]**</code>, ect.
• Each of these tag may only be used once for each scenario</p>
<p>Then <a href="https://github.com/mattwynne">@mattwynne</a> and <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> are suggesting that you use these rules:</p>
<p>• Each scenario must be tagged.
• The tag must be of the format <code>@feature @feature[A-Z]**</code>, <code>@regression @regression[A-Z]**</code>, ect.
• Each of these tag may only be used once for each scenario</p>
<p>However with the latter you are concerned that it will harder to get right.</p>
<p>I think this is a problem that could be solved by a linter. Unfortunately Gherkin currently doesn't have one, but I've been looking for use cases to motivate people to create the necessary infrastructure.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 09:20:08
[cucumber/common] Comment on #1668
<p><a href="https://github.com/mpkorstanje">@mpkorstanje</a> As I said earlier we could absolutely add a <code>@feature</code> tag next to each <code>@featureXYZ</code> tag; it just seems both redundant and fragile.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 10:06:04
[cucumber/common] Comment on #1668
<p>Overal I can see a few problems with expanding the tag expression language.</p>
<ol><li>There is no room in the reserved character set to expand it without a breaking change.</li><li>It only covers a relatively small number of use cases that don't align with the purpose of tags (they are supposed to be composed).</li><li>For more advanced filtering needs a solution that enables programmatic selection of the scenarios to execute would be more effective.</li>
</ol>
<p>Such a solution would be able to handle any arbitrary filtering criteria. It is a bit more difficult to use then a tag expression but that is not unreasonable given the more esoteric needs it covers.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 10:08:50
[cucumber/common] Comment on #1668
<p>Sure. Feel free to close.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 10:27:08
[cucumber/common] Comment on #1668
<p>I'm not too familiar with the api of cucumber-js, it may not be possible yet, so I was waiting for feedback from others.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 10:34:56
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
GitHub
2021-07-20 11:40:33
[cucumber/common] Comment on #1662
<p><a href="https://github.com/mpkorstanje">@mpkorstanje</a> I have a java program that processes gherkin files.<br />
A portion of my java program needs to be able to, given a gherkin string, source format that gherkin. Most importantly so that the Test Data table is formatted pretty.<br />
Can this be done simply?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 11:51:04
[cucumber/common] Comment on #1662
<p>Is there a reason it has to be done in Java?</p>
<p>We currently don't have any such function in Java. We do have a function that does it in Javascript. If it absolutely has to be Java then I guess you could port it. The structure isn't too different.</p>
<p><a href="https://github.com/cucumber/common/blob/main/gherkin-utils/javascript/src/pretty.ts">https://github.com/cucumber/common/blob/main/gherkin-utils/javascript/src/pretty.ts</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 11:51:59
[cucumber/common] Comment on #1662
<p>Yeah unfortunately in my case java is a requirement because my app server is java and there is no js engine available.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 12:57:55
[cucumber/common] Comment on #590
<p>Yes, it seems to be too late to have a discussion about the usefulness of this as it is already there... It is sad though that we made an agreement by considering pro and con arguments, but finally we (including myself, no blaming here) haven't taken care of applying it and finally the power of status quo have overtaken our arguments.</p>
<p>Now everyone thinks over the usefulness of this feature and certainly will find use cases where it is useful. But (and I am just repeating myself), the question is not whether it is useful for <em>you</em>, but whether the complexity and confusion it introduces is overweighted by the benefits for the thousands of the other users. I meet everyday cucumber/specflow users on my courses: 2/3 of them thinks that background executes only once before all scenarios run. These people get a feature file (the ubiquitous language) to review. They will not check the documentation of Gherkin, because they think they understand the background (a nice example of "unknown unknown").</p>
<p>As I said, it is too late for the rule backgrounds now (and probably not that important anyway), but I just wanted to highlight that we have to make different considerations when we change Gherkin than any other parts of cucumber. I think I could also find good use cases where <em>I</em> could safely use variables, loops or conditional steps in Gherkin, but that is not the point here.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-20 13:55:11
[cucumber/common] Comment on #1653
<p>Having slept on that and reading the questions of <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> made me think... Somehow I miss the link of this service to the real usage. Maybe you have had concrete discussions about that, but now I feel that without any concrete plan/need for using it, we might go into a wrong direction. Now I remember, that when I did this for the SpecFlow Visual Studio extension, the real challenge was not to find and collect the usages, but how this can be integrated into a real IDE. And those challenges were not foreseeable. I remember that I was also worried by the memory consumption, but finally the two key challenges were:</p>
<ol><li>How to build the StepDocuments in a reasonable time. (In a larger project parsing through all feature files takes a few seconds, so you definitely cannot do this when the auto-complete is triggered.)</li><li>If you want to cache the StepDocuments (because of the challenge above), how can you (partially) invalidate/update this cache.</li>
</ol>
<p>I have never implemented an LSP server, but probably doing so would bring up a few other questions as well. So probably before we move forward with this, we should really find a concrete project that needs this (maybe Cucumber Studio, maybe something else) and try to integrate it there and see how it works on a project with 500+ feature files.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-21 00:00:40
[cucumber/common] Comment on #1668
<p>I share your opinion <a href="https://github.com/mpkorstanje">@mpkorstanje</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-21 00:05:07
[cucumber/common] Comment on #1662
<p>I think that could be nice to port gherkin-utils to other languages. That's why I've added the "Help wanted" badge :)</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-21 00:11:59
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-21 00:12:59
[cucumber/common] Comment on #1667
<p>If no other objection (<a href="https://github.com/vincent-psarga">@vincent-psarga</a> <a href="https://github.com/aslakhellesoy">@aslakhellesoy</a>) I think we should merge it</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-21 00:19:49
[cucumber/common] Comment on #1665
<p><a href="https://github.com/ehuelsmann">@ehuelsmann</a> is this PR ready to be merged?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-21 00:38:34
[cucumber/common] Comment on #1670
<p>For info, with the Ruby DTOs, it has been chosen to use read-only attributes</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-21 06:04:10
[cucumber/common] Comment on #1665
<p><a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> yes, it is. I was waiting for an answer to the effect on version numbering; <a href="https://github.com/olleolleolle">@olleolleolle</a> explained that upping the minimum version should be only a minor version upgrade. So, lets merge!</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-21 06:10:59
[cucumber/common] Comment on #1665
<p>Merging with the implied consent.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-21 06:11:06
[cucumber/common] Pull request merged by ehuelsmann
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-21 10:50:10
[cucumber/common] Comment on #1662
<p>I could take a stab at it. there's not much code there. But i'm super weak with ts. So it'll take me a while</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-21 11:23:48
[cucumber/common] Comment on #1662
<p>Would be helpfull for IDEs as well!</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-21 17:00:37
[cucumber/common] Comment on #1668
<p>I suggested such a tag-matching mechanism with wildcards (like: <code>tag-starts-with-prefix</code>, <code>tag-ends-with-suffix</code>, <code>tag-contains-part</code>) for cucumber tag expressions a while ago (after providing the functionality in the Python implementation of cucumber tag-expressions). Because the idea was not accepted at the time, I moved the functionality to a slightly extended variant of tag-expressions in <code>behave</code>.</p>
<p>The extra functionality is not that big (but may need some minor refactoring in other language implementations).</p>
<p>RELATED: Tag-Expression parser and model extensions for providing tag-matching with wildcards</p>
<p>• <a href="https://github.com/behave/behave/blob/master/behave/tag_expression/parser.py#L26">behave.tag<em>expression.parser</a>
• <a href="https://github.com/behave/behave/blob/master/behave/tag_expression/model_ext.py#L44">behave.tag</em>expression.model_ext</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-21 22:27:50
[cucumber/common] Comment on #1668
<p>I like your implementation <a href="https://github.com/jenisys">@jenisys</a> - it strikes the balance of simplicity in both the problem and solution domain nicely.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 00:30:43
[cucumber/common] Comment on #1670
<p>This seems to generate things like the following:</p>
<pre><code>Cucumber::Messages::Cucumber::Messages::Location-&gt;new()
</code></pre>
<p>Is this expected?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 02:05:39
[cucumber/common] Comment on #1662
<p>Short reminder: if anyone would need any kind of help, feel free to join our <a href="https://cucumberbdd-slack-invite.herokuapp.com/">community slack</a>. You'll find plenty of developers ready to help.</p>
<p>If for any reason you would not or can not join slack, we will find other way to provide any kind of help. It can be as part of this issue, or a pull-request, or something else? Are you using some tools which you would keep using? We can certainly join you there. Really, feel free to ask! We're here to help 😃</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 02:52:47
[cucumber/common] Comment on #1659
<p>Thanks - I've refactored based on your suggestion above.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 02:54:44
[cucumber/common] Comment on #1670
<p>No. Thansks for spotting that. The serializer/deserializer test doesn't capture this problem, so I'll need to implement more tests :-).</p>
<p>This error is a consequence of having changed the definition of <code>class_name()</code> and not working the consequences fully through. ETOOMUCHLATENIGHTCODING?</p>
<p>I'll fix in a sec.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 02:54:45
[cucumber/common] ehuelsmann commented on aurelien-reeves's pull request
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 06:51:47
[cucumber/common] Pull request opened by aslakhellesoy
<p><strong>Summary</strong></p>
<p>When multiple step definitions use the same parameter type (e.g. <code>{int}</code> or <code>{flight}</code>), the values from all steps should<br />
be added to the same set of options.</p>
<p><strong>Details</strong></p>
<p>See the new <code>### Example: {int} and {word} choices are build from three steps</code> that illustrates this behaviour.</p>
<p><strong>Motivation and Context</strong></p>
<p>When we fill in options for a parameter type, we want as many example values as possible, so that users can choose from all the values that have been used for a particular parameter type.</p>
<p>☐ Bug fix (non-breaking change which fixes an issue).
☐ New feature (non-breaking change which adds functionality).
☑︎ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<p><strong>Checklist:</strong></p>
<p>☐ The change has been ported to Java.
☐ The change has been ported to Ruby.
☑︎ The change has been ported to JavaScript.
☐ The change has been ported to Go.
☐ The change has been ported to .NET.
☐ I've added tests for my code.
☐ My change requires a change to the documentation.
☐ I have updated the documentation accordingly.
☐ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 06:52:14
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 07:17:03
[cucumber/common] Issue closed by mattwynne
<https://github.com/cucumber/common/issues/590|#590 Gherkin: Remove Background from Rule in grammar>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 07:36:17
[cucumber/common] Comment on #1667
<p>> The dependency to cucumber 4.1 in <code>json-to-messages/ruby-testdata</code>: is it still a requirement?</p>
<p>Well, the idea was to use an "old" Cucumber to ensure that we can treat JSON files produced by users who do not want to upgrade (or can not upgrade).<br />
I don't think it's a good idea to upgrade, the goal of <code>json-to-messages</code> is to support old versions of Cucumber, not the new ones that can natively produce messages ;)</p>
<p>I even think that, to avoid having this conversation at each major release of cucumber, we should add this to the WhiteSource configuration :D</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 07:47:15
[cucumber/common] Comment on #590
<p>During the weekly update meeting just a minute ago, we confirmed <a href="https://github.com/gasparnagy">@gasparnagy</a> 's wording followed by the decision to close this issue without further action. Use of backgrounds remains discouraged, but removal is - due to the time it has been available now - no longer an option.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 07:59:43
[cucumber/common] aurelien-reeves approved davidjgoss's pull request
<p>Looks good to me</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 08:00:23
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 08:01:27
[cucumber/common] Comment on #1667
<p>> Well, the idea was to use an "old" Cucumber to ensure that we can treat JSON files produced by users who do not want to upgrade (or can not upgrade).<br />
> I don't think it's a good idea to upgrade, the goal of <code>json-to-messages</code> is to support old versions of Cucumber, not the new ones that can natively produce messages ;)
>
> I even think that, to avoid having this conversation at each major release of cucumber, we should add this to the WhiteSource configuration :D</p>
<p>Good point! Thank you!</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 10:30:31
[cucumber/common] Comment on #1653
<p>It would be great to see a decent auto-complete for Gherkin in VS Code.<br />
Unfortunately I don't have enough time right now to implement using this solution.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 11:34:09
[cucumber/common] Pull request merged by davidjgoss
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 14:53:50
[cucumber/common] Pull request opened by ehuelsmann
<p>Simply add generic configuration to spawn additional Perl subprojects off from. Using a PR because I can't commit directly to <code>main</code>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 14:55:07
[cucumber/common] Comment on #1673
<p>Removing skip ci because the tests need to run before I can merge...</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 15:05:23
[cucumber/common] Comment on #1673
<p>Based on the failure, this was a bad idea. I don't want Makefile, dist.ini or cpanfile to be overwritten.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 15:05:25
[cucumber/common] Pull request closed by ehuelsmann
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 21:22:49
[cucumber/common] Comment on #1653
<p><a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> <a href="https://github.com/gasparnagy">@gasparnagy</a> I have updated the documentation with an <a href="https://github.com/cucumber/common/tree/main/suggest/javascript#architecture">architecture</a> section.</p>
<p>A full suggest system would need an ETL process to build <code>StepDocument</code> objects and store them in persistent storage. This process would not run in the editor - it would be a separate process running in the background. It could be triggered whenever the step definitions or gherkin steps change.</p>
<p>The ETL process is probably the most memory hungry part of the system, since it would need to load all step definitions and gherkin steps into memory. For an online editor, it could run on a server, and would therefore not affect memory consumption on the machine running the editor.</p>
<p>For a desktop editor (such as VS Code) it would have to run on the same machine, but still in a separate process - either in the LSP server, or even as a completely separate process.</p>
<p>To get a rough idea of memory consumption, we could consider a system where there are 1000 feature files with 200 steps in each. Let's assume the average length of a step is 80 bytes - that would amount to 16Mb. Maybe there are 1000 step definitions, each with an average length of 100 bytes - that's another 100Kb (negligible compared to steps).</p>
<p>I think some benchmarks are needed in order to get a sense of how much memory be consumed, as well as how long it would take.</p>
<p>Regarding real usage - I have already built an editor prototype using Monaco Editor and a custom plugin with an in-memory lucene-like index. It seems to work pretty well, but I have not tried with a large index.</p>
<p>Regarding point 1 <a href="https://github.com/gasparnagy">@gasparnagy</a> - I believe that would be addressed by building the index in a background process. There may be a few seconds lag to update the index when new step definitions are added, but I doubt the users would notice - most of the time they would be writing steps using data that is already in the index.</p>
<p>Regarding point 2 I am not sure partial update of the index is needed. As I said, I don't think usability will suffer significantly if the index is only updated sporadically, for example with a debounce of 10 seconds inactivity and max once a minute for example.</p>
<p>Regarding a concrete project that needs this - we are currently planning on using this in Cucumber Studio and/or Cucumber Reports (that's why I started this library), so I expect integrating early versions of this system in those products will quickly give us more feedback. And we'll open source as much as we can (the LSP stuff and/or the Monaco plugin).</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 21:27:11
[cucumber/common] Pull request merged by aslakhellesoy
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 23:52:44
[cucumber/common] Comment on #1670
<p>Caution: there is already a <code><a href="http://default.mk">default.mk</a></code> for perl in the templates.</p>
<p>At some point we will have to move your code from the <code><a href="http://default.mk">default.mk</a></code> you have created into the one in <code>templates/perl</code>, and <code>rsync</code> the files to other modules which have perl implementations.</p>
<p>Edit: I noticed that there may have been an attempt in <a href="https://github.com/cucumber/common/pull/1673">#1673</a>?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-22 23:55:38
[cucumber/common] Comment on #1653
<p>Thanks for all those explanation <a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> 👍</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-23 01:46:18
[cucumber/common] Pull request opened by ehuelsmann
<p>Factor out common build infrastructure for Perl projects in order to be able to share it later with the <code>messages-perl</code> implementation.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-23 02:39:39
[cucumber/common] Pull request merged by ehuelsmann
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-23 02:58:20
[cucumber/common] Comment on #1670
<p>> Caution: there is already a <code><a href="http://default.mk">default.mk</a></code> for perl in the templates.
>
> At some point we will have to move your code from the <code><a href="http://default.mk">default.mk</a></code> you have created into the one in <code>templates/perl</code>, and <code>rsync</code> the files to other modules which have perl implementations.
>
> Edit: I noticed that there may have been an attempt in <a href="https://github.com/cucumber/common/pull/1673">#1673</a>?</p>
<p>Yes, that was an attempt, based on misunderstanding the situation. I've since merged a better version and rebased this branch. By setting up a '.rsync' file to ensure the shared infrastructure remains current, this new library should now be integrated in the regular build infrastructure.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-23 02:59:58
[cucumber/common] Pull request ready for review by ehuelsmann
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-23 03:52:13
[cucumber/common] Pull request opened by ehuelsmann
<p>Copy the 'release-includes-unversioned artifacts' risk mitigation<br />
to gherkin-perl to get consistent packaging.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-23 06:18:06
[cucumber/common] Pull request merged by ehuelsmann
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-23 11:12:39
[cucumber/common] Comment on #1670
<p>I think we might need to change this to something else that represents to Cucumber Open community.</p>
<p>We’ve been using Cucumber Ltd for other repos, but that company is defunkt. I’m not sure we ahould use SmartBear either, since it might deter people from contributing.</p>
<p><a href="https://github.com/mattwynne">@mattwynne</a> <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a> what do you think? Time to revisit the license owner of all the cucumber/** repos?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-23 11:12:41
[cucumber/common] aslakhellesoy requested changes on aurelien-reeves's pull request
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-23 13:31:40
[cucumber/common] Comment on #1670
<p>Let me know what it should be. if there were a legal body, we would be able to work with ICLA or CCLA constructs like the Apache Foundation does. That would be the most sound solution. I can change it to "The Cucumber Project", but maybe that's just an empty shell/meaningless term at the moment?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-23 14:12:28
[cucumber/common] Comment on #1670
<p><a href="https://github.com/mattwynne">@mattwynne</a> <a href="https://github.com/aslakhellesoy">@aslakhellesoy</a> maybe it's consistent for now to (at least) add Cucumber Ltd for now? That way, it's consistent with the other repositories.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-23 14:19:09
[cucumber/common] Comment on #1670
<p>That would be great - thanks. Then we can address the license owner issue on a broader scale later.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-23 14:24:40
[cucumber/common] Comment on #1670
<p>Ok. I actually already set the copyright holder in the metadata to Cucumber Ltd: <a href="https://github.com/cucumber/common/pull/1670/files#diff-e4960b209d869c770dab6dae20e4e07a5823e193b23d01967012993cde5c907eR12">https://github.com/cucumber/common/pull/1670/files#diff-e4960b209d869c770dab6dae20e4e07a5823e193b23d01967012993cde5c907eR12</a> I'll submit an additional commit to add Cucumber Ltd to the author/copyright in the generated messages file.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-23 14:28:47
[cucumber/common] Comment on #1670
<p>How about this instead:</p>
<pre><code>=head1 LICENSE
Please see the included LICENSE for the canonical version. In summary:
The MIT License (MIT)
Copyright (c) 2021 Erik Huelsmann
Copyright (c) 2021 Cucumber Ltd
This work is loosely derived from prior work of the same library for Ruby,
called C&lt;cucumber-messages&gt;.
=cut
</code></pre>
<p>?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-23 14:39:17
[cucumber/common] Comment on #1670
<p>I guess my question is: this seems to be in line with the Gherkin/Perl module; maybe that was added in another era though?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-23 14:39:18
[cucumber/common] ehuelsmann commented on aurelien-reeves's pull request
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-23 16:25:54
[cucumber/common] Comment on #1670
<p>Cucumber Ltd is wholly owned by SmartBear, so really they're interchangeable. I can't comment on whether it would deter people or not.</p>
<p>I don't really understand what it means to copyright something with an MIT license anyway. Does it enable you to sue someone if they tried to claim they made it instead of you?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-24 00:59:52
[cucumber/common] Comment on #1670
<p>As the copyright holder, you're allowed to change the license as you own the creative rights to the code. With the MIT license, the distributor and user can do with it what they want (even create derived works, or incorporate the code in commercial products), but they can't change the license of the original code without the consent of the original copyright holders. Also, the copyright holders are the ones <em>giving</em> the MIT consent. For a comparison of licenses and summary of rights by license: <a href="https://en.wikipedia.org/wiki/Comparison_of_free_and_open-source_software_licences#FOSS_licences">https://en.wikipedia.org/wiki/Comparison<em>of</em>free<em>and</em>open-source<em>software</em>licences#FOSS_licences</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-24 01:51:13
[cucumber/common] Comment on #1670
<p>The code on which this implementation is loosely based (the ruby messages implementation), lists an MIT license with Cucumber Ltd copyrights.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-26 04:30:08
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-26 08:26:18
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-27 00:15:28
[cucumber/common] Issue opened by tim-yao
<p><strong>Is your feature request related to a problem? Please describe.</strong><br />
Cucumber report(html, json formatter) doesn't include retried details when using <code>--retry</code>.</p>
<p>When we use <code>--retry</code>, report will mark the flaky test pass if there is one retry get pass. That is working as expected. However, if we can also add some details of <code>retry</code> in the report, e.g show the number of failed attempt, that will help us to find potential intermittent bugs, instead of hiding bugs.</p>
<p>The idea is coming from <a href="https://automationpanda.com/2021/06/14/are-automated-test-retries-good-or-bad/">https://automationpanda.com/2021/06/14/are-automated-test-retries-good-or-bad/</a></p>
<p><strong>Describe the solution you'd like</strong><br />
Something on the report to indicate there are retry attempts happened. Like the number of failed attempt. A example:<br />
<a href="https://user-images.githubusercontent.com/31641325/126961755-ac5ac219-2b6f-4d04-adb4-e3cc94d22fc3.png">specflow-report</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-27 00:18:51
[cucumber/common] Comment on #1677
<p>I think the messages would have all the info for all messages-based formatters to be enhanced too</p>
<p>But first, let's focus on the html-formatter</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-28 11:04:38
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-28 11:13:39
[cucumber/common] Comment on #1630
<p>OK! Dart is now in <a href="https://github.com/cucumber/build/blob/main/CHANGELOG.md#060---2021-07-28">v6.0.0 of the build image</a>.</p>
<p>The robots have already created <a href="https://github.com/cucumber/common/pull/1678">#1678</a> which upgrades the <code>main</code> branch to use the new build image. If that passes, we can merge it, rebase this branch and it should then be running with Dart in the build image.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-28 11:29:31
[cucumber/common] Comment on #1646
<p>What about caching <code>make</code> artefacts using <a href="https://github.com/actions/cache">https://github.com/actions/cache</a>? Could that work? Maybe we could do <code>make clean</code> in a nightly build or something?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-29 00:26:37
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-29 00:28:24
[cucumber/common] Comment on #1646
<p>At the moment the CI is using CircleCI, not github actions.<br />
But they certainly have some cache mechanism too 🤔</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-29 02:31:22
[cucumber/common] Comment on #1646
<p>That is putting a lot of trust in the correctness of our make files and once broken, almost impossible to debug.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-07-29 02:40:50
[cucumber/common] Comment on #1646
<p>Yes. I must say that I like the serial build being totally fresh each times. Using a cache, and it is already not idempotent anymore</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-01 15:01:20
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
rehypejs/rehype-sanitize
<strong><a href="https://togithub.com/rehypejs/rehype-sanitize/releases/5.0.0"><code>v5.0.0</code></a></strong></p>
<p><a href="https://togithub.com/rehypejs/rehype-sanitize/compare/4.0.0...5.0.0">Compare Source</a></p>
<p>• <a href="https://togithub.com/rehypejs/rehype-sanitize/commit/abf5e36"><code>abf5e36</code></a> Add JSDoc based types
• <a href="https://togithub.com/rehypejs/rehype-sanitize/commit/c0f5518"><code>c0f5518</code></a> Use ESM
• Change:
<pre><code>// From CommonJS
var rehypeSanitize = require('rehype-sanitize')
// To ESM
import rehypeSanitize from 'rehype-sanitize'
</code></pre></p>
<pre><code> Learn <a href="https://gist.github.com/sindresorhus/a39789f98801d908bbc7ff3ecc99d99c">more about ESM in this guide</a>
</code></pre>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-02 01:32:20
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
rehypejs/rehype-raw
<strong><a href="https://togithub.com/rehypejs/rehype-raw/releases/6.0.0"><code>v6.0.0</code></a></strong></p>
<p><a href="https://togithub.com/rehypejs/rehype-raw/compare/5.1.0...6.0.0">Compare Source</a></p>
<p>• <a href="https://togithub.com/rehypejs/rehype-raw/commit/75258b8"><code>75258b8</code></a> Use ESM
• Change:
<pre><code>// From CommonJS
var rehypeRaw = require('rehype-raw')
// To ESM
import rehypeRaw from 'rehype-raw'
</code></pre></p>
<pre><code> Learn <a href="https://gist.github.com/sindresorhus/a39789f98801d908bbc7ff3ecc99d99c">more about ESM in this guide</a>
</code></pre>
<p>• <a href="https://togithub.com/rehypejs/rehype-raw/commit/1af7117"><code>1af7117</code></a> Add JSDoc based types
• <a href="https://togithub.com/rehypejs/rehype-raw/commit/d361d2d"><code>d361d2d</code></a> Update dependencies</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-03 00:31:34
[cucumber/common] Comment on #1680
<p><a href="https://github.com/rehypejs/rehype/discussions/63">rehypejs/rehype#63</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-03 00:31:56
[cucumber/common] Comment on #1679
<p><a href="https://github.com/rehypejs/rehype/discussions/63">rehypejs/rehype#63</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-04 03:01:05
[cucumber/common] Issue opened by s-burtonshaw
<p><strong>Describe the bug</strong><br />
The executable <https://github.com/cucumber/common/releases/download/cucumber-json-formatter%2Fv7.0.1/cucumber-json-formatter-windows-amd64|cucumber-json-formatter-windows-amd64 v7.01> is being detected as Trojan named Artemis!2DD068390E80 by McAfee Antivirus software.</p>
<p><strong>To Reproduce</strong><br />
Steps to reproduce the behavior:</p>
<ol><li>On Windows system with McAfee antivirus attempt to download latest version of cucumber-json-formatter</li><li>Attempt to use/rename exe</li><li>McAfee on-access scan detects trojan (and deletes exe)</li>
</ol>
<p><strong>Expected behavior</strong><br />
No detection.</p>
<p><strong>Context & Motivation</strong><br />
We are using cucumber-json-formatter in our AzureDevops release pipeline to generate a HTML report. We are using it to convert ndjson results from cucumber to JSON to be consumed by <a href="https://marketplace.visualstudio.com/items?itemName=MaciejMaciejewski.azure-pipelines-cucumber&ssr=false#overview|this">https://marketplace.visualstudio.com/items?itemName=MaciejMaciejewski.azure-pipelines-cucumber&ssr=false#overview|this</a> extension which in turn uses <a href="https://www.npmjs.com/package/cucumber-html-reporter">this</a> node package.</p>
<p>We are using a previous version of cucumber-json-formatter (unsure what version as I can't see any version info on the exe) and this is not detected by McAfee but wanted to try the latest version I was trying to get screenshots to show on failed steps (not even sure this new version will help with that problem yet)</p>
<p><strong>Screenshots</strong><br />
<a href="https://user-images.githubusercontent.com/10601691/128153056-6159a62d-6d32-4691-877d-b87464a39ef4.png">image</a></p>
<p><strong>Your Environment</strong></p>
<p>• cucumber-json-formatter-windows-amd64 v7.01
• McAfee Agent - v5.7.3.245
• Windows 10 x64 - v1909</p>
<p>I will start looking at alternative for generating the HTML report but thought this was worth mentioning.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-04 14:55:20
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-05 00:59:30
[cucumber/common] Comment on #1681
<p>Hi, and thanks for your report.</p>
<p>This is because that standalone json-formatter is written in Go.<br />
Go often triggers false-positives on Windows. Refs. <a href="https://golang.org/doc/faq#virus">https://golang.org/doc/faq#virus</a></p>
<p>The only thing that could be done may be to send a report to McAfee.</p>
<p>Please note that we have recently decided to undeprecate built-in json formatters. We won't enhanced them, but we won't remove them either. That means that you can actually rely on the report that <code>--format json</code> will generate.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-05 01:04:23
[cucumber/common] Pull request merged by aurelien-reeves
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-05 07:25:17
[cucumber/common] Comment on #1670
<p>I think this is good enough for now.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-05 07:48:28
[cucumber/common] Comment on #768
<p>I really don't like this suggestion:</p>
<p>• IMO it encourages ambiguous business domain language, with automation code modified by preceding keyword
• Gherkin currently supports steps being preceded by '**'
• when conjunctions are used (And, But) the reader will have to identify the preceding G/W/T keyword</p>
<p>However, a number of Gherkin implementations do work in this way (SpecFlow, Perl, Behat?). SpecFlow introduced the @StepDefinition annotation to emulate core Cucumber implementations.</p>
<p>I support the passing of the keyword to the pickle.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-05 07:57:33
[cucumber/common] Comment on #768
<p>I also support passing of the keyword to the pickle.</p>
<p>But I do not support making the keyword part of the step definition. Major breaking change with no value.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-05 08:57:00
[cucumber/common] Comment on #1630
<p><a href="https://github.com/lamkr">@lamkr</a> could you rebase this so we can get it merged in?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-05 12:55:17
[cucumber/common] Pull request opened by renovate[bot]
<https://github.com/cucumber/common/pull/1683|#1683 Update module github.com/onsi/gomega to v1.15.0>
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
onsi/gomega
<strong><a href="https://togithub.com/onsi/gomega/releases/v1.15.0"><code>v1.15.0</code></a></strong></p>
<p><a href="https://togithub.com/onsi/gomega/compare/v1.14.0...v1.15.0">Compare Source</a></p>
<p><strong>1.15.0</strong>
<strong>Fixes</strong></p>
<p>The previous version (1.14.0) introduced a change to allow <code>Eventually</code> and <code>Consistently</code> to support functions that make assertions. This was accomplished by overriding the global fail handler when running the callbacks passed to <code>Eventually/Consistently</code> in order to capture any resulting errors. Issue <a href="https://togithub.com/onsi/gomega/issues/457">#457</a> uncovered a flaw with this approach: when multiple <code>Eventually</code>s are running concurrently they race when overriding the singleton global fail handler.</p>
<p>1.15.0 resolves this by requiring users who want to make assertions in <code>Eventually/Consistently</code> call backs to explicitly pass in a function that takes a <code>Gomega</code> as an argument. The passed-in <code>Gomega</code> instance can be used to make assertions. Any failures will cause <code>Eventually</code> to retry the callback. This cleaner interface avoids the issue of swapping out globals but comes at the cost of changing the contract introduced in v1.14.0. As such 1.15.0 introduces a breaking change with respect to 1.14.0 - however we expect that adoption of this feature in 1.14.0 remains limited.</p>
<p>In addition, 1.15.0 cleans up some of Gomega's internals. Most users shouldn't notice any differences stemming from the refactoring that was made.</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-05 14:12:08
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
cucumber/cucumber-jvm
<strong><a href="https://togithub.com/cucumber/cucumber-jvm/blob/master/CHANGELOG.md#%E2%80%8B6110-2021-08-05|<code">https://togithub.com/cucumber/cucumber-jvm/blob/master/CHANGELOG.md#%E2%80%8B6110-2021-08-05|<code</a>v6.11.0</code>></strong></p>
<p><a href="https://togithub.com/cucumber/cucumber-jvm/compare/v6.10.4...v6.11.0">Compare Source</a></p>
<p><strong>Added</strong></p>
<p>• [TestNG] Add <code>CucumberPropertiesProvider</code> to allow properties from <code>testng.xml</code> to be used (<a href="https://togithub.com/cucumber/cucumber-jvm/pull/2351">#2351</a> Gayan Sandaruwan)</p>
<p><strong>Changed</strong></p>
<p>• [JUnit Platform] Update dependency org.junit.platform:junit-platform-engine to v1.7.2
• [Core] Update dependency org.apiguardian:apiguardian-api to v1.1.2</p>
<p><strong>Fixed</strong></p>
<p>• [Core] Fixed reports banner to point to <a href="https://cucumber.io/docs/cucumber/environment-variables/">new docs</a> about environment variables
• [Core] Remove <code>--add-plugin</code> alternate name from USAGE.txt (<a href="https://togithub.com/cucumber/cucumber-jvm/pull/2319">#2319</a> ebreck)</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about these updates again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-05 14:43:51
[cucumber/common] Issue opened by mpkorstanje
<https://github.com/cucumber/common/issues/1685|#1685 Inconsistent copyright attribution in license>
<p>During the contributors meeting we observed that the copyright attribution in the various license files is inconsistent between projects and between reality.</p>
<p>I would suggest using "Cucumber Ltd and Contributors".</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-05 14:57:12
[cucumber/common] Pull request merged by mpkorstanje
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-06 00:50:24
[cucumber/common] Comment on #1670
<p>As <a href="https://github.com/cucumber/common/pull/1670#discussion_r683503958">#1670 (comment)</a> and as discussed during our last community meeting: this is good for now as part of this PR.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-06 00:58:42
[cucumber/common] Comment on #1685
<p>As part of SmartBear, I have to suggest to also put "SmartBear" in the copyright. I am pretty sure we will have to, in addition or in replacement of the "Cucumber Ltd" one.</p>
<p>However, I would encourage putting the contributors first.</p>
<p>Maybe we may also consider having an explicit list of contributors, in a dedicated file (CONTRIBUTORS.md?) listing all contributors - even people who have contributed a single time - to the code? Git would then help to associate contributors to line of codes if this would be required later.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-06 00:59:34
[cucumber/common] Pull request merged by aurelien-reeves
<https://github.com/cucumber/common/pull/1683|#1683 Update module github.com/onsi/gomega to v1.15.0>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-06 01:02:05
[cucumber/common] Comment on #1679
<p>From there: <a href="https://github.com/rehypejs/rehype/discussions/63#discussioncomment-1123797">rehypejs/rehype#63 (comment)</a></p>
<p>> Perhaps with a <code>@ts-expect-error</code> for now — the actual code should just work but the types depend on a new major of unified (which is already released but isn’t in <code>react-markdown</code> yet)</p>
<p>As there is no release plan for now, we may wait for an update of <code>react-markdown</code> before considering ignoring that error with <code>@ts-expect-error</code>?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-06 01:02:33
[cucumber/common] Comment on #1680
<p>From there: <a href="https://github.com/rehypejs/rehype/discussions/63#discussioncomment-1123797">rehypejs/rehype#63 (comment)</a></p>
<p>> Perhaps with a <code>@ts-expect-error</code> for now — the actual code should just work but the types depend on a new major of unified (which is already released but isn’t in <code>react-markdown</code> yet)</p>
<p>As there is no release plan for now, we may wait for an update of <code>react-markdown</code> before considering ignoring that error with <code>@ts-expect-error</code>?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-06 01:34:10
[cucumber/common] Comment on #1685
<p>> Maybe we may also consider having an explicit list of contributors, in a dedicated file (CONTRIBUTORS.md?) listing all contributors - even people who have contributed a single time - to the code?</p>
<p>I'd consider that another issue that can be resolved separately from or after this one.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-06 04:15:59
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
remarkjs/remark-gfm
<strong><a href="https://togithub.com/remarkjs/remark-gfm/releases/2.0.0"><code>v2.0.0</code></a></strong></p>
<p><a href="https://togithub.com/remarkjs/remark-gfm/compare/1.0.0...2.0.0">Compare Source</a></p>
<p>• <a href="https://togithub.com/remarkjs/remark-gfm/commit/6c518d8"><code>6c518d8</code></a> Use ESM
<pre><code>// From CommonJS
var remarkGfm = require('remark-gfm')
// To ESM
import remarkGfm from 'remark-gfm'
</code></pre></p>
<pre><code>Learn <a href="https://gist.github.com/sindresorhus/a39789f98801d908bbc7ff3ecc99d99c">more about ESM in this guide</a>
</code></pre>
<p>• <a href="https://togithub.com/remarkjs/remark-gfm/commit/37d92ac"><code>37d92ac</code></a> Add JSDoc based types
• <a href="https://togithub.com/remarkjs/remark-gfm/commit/0846014"><code>0846014</code></a> Remove warning for remark 12</p>
<p><strong>Full Changelog</strong>: <a href="https://github.com/remarkjs/remark-gfm/compare/1.0.0...2.0.0">remarkjs/remark-gfm@1.0.0...2.0.0</a></p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-06 12:21:06
[cucumber/common] Comment on #1630
<p>Hi <a href="https://github.com/lamkr">@lamkr</a>,</p>
<p>Thanks for your making your first contribution to Cucumber, and welcome to the Cucumber committers team! You can now push directly to this repo and all other repos under the cucumber organization! 🍾</p>
<p>In return for this generous offer we hope you will:</p>
<p>• ✅ Continue to use branches and pull requests. When someone on the core team approves a pull request (yours or someone else's), you're welcome to merge it yourself.
• 💚 Commit to setting a good example by following and upholding our <a href="https://cucumber.io/conduct">code of conduct</a> in your interactions with other collaborators and users.
• 💬 Join the <a href="http://cucumber.io/support#slack">community Slack channel</a> to meet the rest of the team and make yourself at home.
• ℹ️ Don't feel obliged to help, just do what you can if you have the time and the energy.
• 🙋 Ask if you need anything. We're looking for feedback about how to make the project more welcoming, so please tell us!</p>
<p>On behalf of the Cucumber core team,<br />
Aslak Hellesøy<br />
Creator of Cucumber</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-06 12:21:06
[cucumber/common] Pull request merged by mattwynne
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-07 02:47:51
[cucumber/common] Issue opened by jamietanna
<p><strong>Describe the bug</strong></p>
<p>There is a typo in <code>Token.java</code>'s <code>mathchedItems</code>, which should be <code>matchedItems</code></p>
<p><strong>To Reproduce</strong></p>
<p>See</p>
<p><a href="https://github.com/cucumber/common/blob/b94ce625967581de78d0fc32d84c35b46aa5a075/gherkin/java/src/main/java/io/cucumber/gherkin/Token.java#L10">common/gherkin/java/src/main/java/io/cucumber/gherkin/Token.java</a></p>
<p>Line 10 in </cucumber/common/commit/b94ce625967581de78d0fc32d84c35b46aa5a075|b94ce62></p>
<p>is currently <code>mathched</code> not <code>matched</code></p>
<p><strong>Expected behavior</strong></p>
<p>Typo is corrected, at our next major version bump of Gherkin.</p>
<p><strong>Context & Motivation</strong></p>
<p>Can be confusing to read.</p>
<p><strong>Screenshots</strong></p>
<p>N/A</p>
<p><strong>Your Environment</strong></p>
<p>Gherkin v20.0.1</p>
<p><strong>Additional context</strong></p>
<p>N/A</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-08 06:57:51
[cucumber/common] Comment on #768
<p>Discussed on the weekly Cucumber team call; consensus was that since some implementations have had the step-definition-per-keyword behavior (mentioned during the meeting: SpecFlow and Test::BDD::Cucumber (Perl)), it's necessary to add the keywords to the pickles. Additionally, since Cucumber never exhibited this behavior, it's expected to be a major breaking change for all those who have Cucumber-based test suites. In other words: Let's do this in the pickle so more projects can move to the generic parser infrastructure, but lets not incorporate this any further than that.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-08 06:59:36
[cucumber/common] Comment on #768
<p>It's my action to implement this in the Gherkin (and Messages) libraries up to the Pickle (for as many languages as I can get).</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-09 00:18:39
[cucumber/common] Issue opened by mattwynne
<p>Using this issue to bring together the efforts scattered across other repos.</p>
<p>The goal is to have all releases to RubyGems, NPM, Maven etc. made by GitHub Actions workflows. We'll use GitHub Environments to secure the secrets used to publish packages.</p>
<p>☑︎ cucumber/build <a href="https://github.com/cucumber/build/issues/42">cucumber/build#42</a>
☑︎ cucumber/cucumber-ruby-wire <a href="https://github.com/cucumber/cucumber-ruby-wire/pull/49">cucumber/cucumber-ruby-wire#49</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-10 14:00:49
[cucumber/common] Comment on #1668
<p><a href="https://github.com/customcommander">@customcommander</a> despite the objections you've heard, I think it would be interesting to see a PR for this, to bring the same functionality Jens has added to Behave into Cucumber. It might turn out that it's not as complicated as people fear. Would you be interested in giving it a go?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-10 23:51:27
[cucumber/common] Comment on #1688
<p>Today I factored out a <a href="https://github.com/cucumber-actions/create-release-pr/">https://github.com/cucumber-actions/create-release-pr/</a> action which will cut a lot of the boilerplate out of adding a <code>pre-release.yaml</code> workflow to a new project.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-11 00:12:07
[cucumber/common] Issue opened by aslakhellesoy
<p>The new gherkin/dart module is not built in CI.</p>
<p>When <a href="https://github.com/cucumber/common/pull/1630">#1630</a> was merged we forgot to add a job to <code>.circleci/config.yml</code>, so it isn't built as a separate job. There is nothing in <code>gherkin/Makefile</code> either, so the module isn't built as part of the serial job either.</p>
<p>/cc <a href="https://github.com/mattwynne">@mattwynne</a> <a href="https://github.com/aurelien-reeves">@aurelien-reeves</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-12 02:29:00
[cucumber/common] Comment on #1668
<p><a href="https://github.com/mattwynne">@mattwynne</a> Oh nice! I'd like to give it a go but:</p>
<ol><li>ETA for a PR: ± end of August at best.</li><li>I can only do JS (or TS) so I wouldn't be able to replicate for different languages.</li>
</ol>
<p>Is that ok?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-12 06:32:14
[cucumber/common] Comment on #1668
<p><a href="https://github.com/mattwynne">@mattwynne</a> it's not about the complexity of the solution, it's about the relative complexity of the solution vs. the gains it provides vs the complexity of alternative solutions.</p>
<p>For example achieving these things with JUnit 5, is easy:</p>
<pre><code>private static class MyPostDiscoveryFilter implements PostDiscoveryFilter {
@Override
public FilterResult apply(TestDescriptor testDescriptor) {
return FilterResult.includedIf(
testDescriptor.isTest() &amp;&amp;
testDescriptor.getTags().stream()
.map(TestTag::getName)
.anyMatch(tag -&gt; tag.startsWith("prefix-")));
}
}
</code></pre>
<p>As is it with TestNG</p>
<pre><code> @Test(groups = "cucumber", description = "Runs Cucumber Scenarios", dataProvider = "scenarios")
public void runScenario(PickleWrapper pickleWrapper, FeatureWrapper featureWrapper) {
if (pickleWrapper.getPickle().getTags().stream().noneMatch(tag -&gt; tag.startsWith("prefix-"))) {
throw new SkipException("Tags didn't match");
}
testNGCucumberRunner.runScenario(pickleWrapper.getPickle());
}
</code></pre>
<p>Cucumber is huge multi-language project already. We should not reinvent the wheel when the test frameworks we integrate with already provide functionality that allows us to do these things.</p>
<p>Now perhaps our integration with javascript test frameworks is such that we currently can't do this programmatically, but I then first suggest looking into improving those integrations rather then adding more functionality to Cucumber. There is no point in reinventing the wheel, just because it was not invented here and we didn't bother to look elsewhere.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-12 06:55:05
[cucumber/common] Comment on #1690
<p>Oops. I figured the fact that the CI was failing before we added Dart to the build image meant it was being tested in CI. I should have looked closer.</p>
<p><a href="https://github.com/lamkr">@lamkr</a> are you interested in helping with this? It would be good to see how easy the CI / Make infrastructure is to understand for someone new to it.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-12 07:13:45
[cucumber/common] Comment on #1685
<p>For reference, Swagger (anther SmartBear open source project) uses "SmartBear Software Inc." <a href="https://github.com/swagger-api/swagger-ui/blob/master/LICENSE">https://github.com/swagger-api/swagger-ui/blob/master/LICENSE</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-12 07:30:17
[cucumber/common] Issue opened by luke-hill
<p>Previously in cucumber 5.3 -> cucumber-html-formatter 9 this functionality worked<br />
Currently in cucumber 7 -> cucumber-html-formatter 16 this functionality is no longer working</p>
<p><strong>Describe the bug</strong><br />
Filter-by passed seems to pull in additional test cases</p>
<p><strong>To Reproduce</strong><br />
Steps to reproduce the behavior:<br />
I have a report that has the following<br />
passed 128 scenarios<br />
skipped 24 scenarios<br />
unknown 59 scenarios</p>
<p><strong>Expected behavior</strong><br />
Filtering works returns only the ones checked</p>
<p><strong>Screenshots</strong><br />
Can pair and show video</p>
<p><strong>Desktop (please complete the following information):</strong></p>
<pre><code>Implementation cucumber-ruby - 7.0.0
Runtime ruby - 2.7.3
OS linux - cucumber/cucumber-react#1 SMP Tue Jun 15 20:57:50 UTC 2021
CPU x86_64
</code></pre>
<p><strong>Additional context</strong><br />
In console we get one warning</p>
<pre><code>Audit usage of navigator.userAgent, navigator.appVersion, and navigator.platform
</code></pre>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-12 11:27:59
[cucumber/common] Comment on #1668
<p>Hey <a href="https://github.com/customcommander">@customcommander</a>,</p>
<p>We talked about this at the <a href="https://docs.google.com/document/d/1ewOKjaK5F17pwv7wtiaBFWCdzHQBXqQMCxqkwWKVmX8/edit#">weekly community Zoom</a> today, and there's still some concern that the benefit of this feature won't justify the ongoing maintenance cost, especially given that this is in a shared component where we'd need to implement in multiple langauges.</p>
<p>With that said, I personally think this conversation could be moved forward by looking at some code, or at least fleshing out some test cases / examples to understand the details of how it would behave in edge-cases if we started supporting some kind of pattern matching (or full blown-regexen).</p>
<p>If you have the energy to do that, given the risk that we might still not achieve consensus to merge this behaviour in, then yes we can of course wait until September or even longer 😄. I think that, if we regard this as a prototype, then implementing in one language first makes sense anyway, so you only being able to implement in JS/TS is fine for now.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-13 02:23:41
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
fluentassertions/fluentassertions
<strong><a href="https://togithub.com/fluentassertions/fluentassertions/compare/5.10.3...6.0.0"><code>v6.0.0</code></a></strong></p>
<p><a href="https://togithub.com/fluentassertions/fluentassertions/compare/5.10.3...6.0.0">Compare Source</a></p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-13 03:59:41
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
remarkjs/react-markdown
<strong><a href="https://togithub.com/remarkjs/react-markdown/blob/master/changelog.md#%E2%80%8B700---2021-08-13|<code">https://togithub.com/remarkjs/react-markdown/blob/master/changelog.md#%E2%80%8B700---2021-08-13|<code</a>v7.0.0</code>></strong></p>
<p><a href="https://togithub.com/remarkjs/react-markdown/compare/6.0.3...7.0.0">Compare Source</a></p>
<p>Welcome to version 7.<br />
This a major release and therefore contains breaking changes.</p>
<p><strong>Breaking changes</strong></p>
<p>• <a href="https://togithub.com/remarkjs/react-markdown/commit/01b11fe"><code>01b11fe</code></a><br />
<a href="https://togithub.com/remarkjs/react-markdown/commit/c613efd"><code>c613efd</code></a><br />
<a href="https://togithub.com/remarkjs/react-markdown/commit/a1e1c3f"><code>a1e1c3f</code></a><br />
<a href="https://togithub.com/remarkjs/react-markdown/commit/aeee9ac"><code>aeee9ac</code></a><br />
Use ESM<br />
(please <a href="https://gist.github.com/sindresorhus/a39789f98801d908bbc7ff3ecc99d99c">read this</a>)
• <a href="https://togithub.com/remarkjs/react-markdown/commit/3dffd6a"><code>3dffd6a</code></a><br />
Update dependencies<br />
(upgrade all your plugins and this should go fine)</p>
<p><strong>Internals</strong></p>
<p>• <a href="https://togithub.com/remarkjs/react-markdown/commit/8b5481c"><code>8b5481c</code></a><br />
<a href="https://togithub.com/remarkjs/react-markdown/commit/fb1b512"><code>fb1b512</code></a><br />
<a href="https://togithub.com/remarkjs/react-markdown/commit/144af79"><code>144af79</code></a><br />
Replace <code>jest</code> with <code>uvu</code>
• <a href="https://togithub.com/remarkjs/react-markdown/commit/8c572df"><code>8c572df</code></a><br />
Replace <code>rollup</code> with <code>esbuild</code>
• <a href="https://togithub.com/remarkjs/react-markdown/commit/8737eac"><code>8737eac</code></a><br />
<a href="https://togithub.com/remarkjs/react-markdown/commit/28d4c75"><code>28d4c75</code></a><br />
<a href="https://togithub.com/remarkjs/react-markdown/commit/b2dd046"><code>b2dd046</code></a><br />
Refactor code-style</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-13 07:01:18
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-13 08:40:25
[cucumber/common] Comment on #1690
<p>> Oops. I figured the fact that the CI was failing before we added Dart to the build image meant it was being tested in CI. I should have looked closer.
>
> <a href="https://github.com/lamkr">@lamkr</a> are you interested in helping with this? It would be good to see how easy the CI / Make infrastructure is to understand for someone new to it.</p>
<p>hi, <a href="https://github.com/mattwynne">@mattwynne</a> ! I don't have that much time available but I will do my best to help. I'm not an expert in CI either, but where can I start?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-13 10:06:48
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
jsdom/jsdom
<strong><a href="https://togithub.com/jsdom/jsdom/blob/master/Changelog.md#%E2%80%8B1700|<code">https://togithub.com/jsdom/jsdom/blob/master/Changelog.md#%E2%80%8B1700|<code</a>v17.0.0</code>></strong></p>
<p><a href="https://togithub.com/jsdom/jsdom/compare/16.7.0...17.0.0">Compare Source</a></p>
<p>Breaking change: Node v12 is now the minimum supported version.</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-13 12:03:23
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-16 07:13:43
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Release Notes</strong>
microsoft/vstest
<strong><a href="https://togithub.com/microsoft/vstest/releases/v16.11.0"><code>v16.11.0</code></a></strong></p>
<p>See the release notes <a href="https://togithub.com/microsoft/vstest-docs/blob/main/docs/releases.md#%E2%80%8B16110|here">https://togithub.com/microsoft/vstest-docs/blob/main/docs/releases.md#%E2%80%8B16110|here</a>.</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-16 07:14:19
[cucumber/common] Pull request closed by renovate[bot]
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-16 09:07:53
[cucumber/common] Comment on #1690
<p><a href="https://github.com/lamkr">@lamkr</a> I think the place to start is to add a <code>gherkin/dart/Makefile</code> with a <code>default</code> target that compiles and tests the code. Then we can integrate that into the main <code>gherkin/Makefile</code> and finally add a job to CI to run it.</p>
<p><a href="https://github.com/ciaranmcnulty">@ciaranmcnulty</a> it strikes me this could present a good opportunity to try our ideas about breaking up the Docker image monolith, since the Makefile here is essentially green field?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-16 11:10:51
[cucumber/common] Comment on #1690
<p>I've realised that the <code>dart</code> command wasn't even in the <code>PATH</code> in the build image, because I hadn't tested the installation script properly. I'll release a new version of the build image soon.</p>
<p>See <a href="https://github.com/cucumber/build/pull/53">cucumber/build#53</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-16 11:34:40
[cucumber/common] Pull request opened by mattwynne
<p><strong>Summary</strong></p>
<p>This means we have a single source of truth for the build image version.</p>
<p><strong>Details</strong></p>
<p>It's been duplicated in the CircleCI config and the Makefile, and the Makefile had drifted out of date. Now we're reading the value to use in the Makefile from the CircleCI config, so we only have to change it in on place.</p>
<p><strong>How Has This Been Tested?</strong></p>
<p>I ran <code>make docker-run</code></p>
<p><strong>Types of changes</strong></p>
<p>☑︎ Refactoring / tech debt / DX improvements
☐ Bug fix (non-breaking change which fixes an issue).
☐ New feature (non-breaking change which adds functionality).
☐ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-16 13:54:35
[cucumber/common] Pull request merged by mattwynne
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-16 14:50:23
[cucumber/common] Pull request opened by renovate[bot]
<p><a href="https://renovatebot.com">WhiteSource Renovate</a></p>
<p>This PR contains the following updates:</p>
<hr />
<p><strong>Configuration</strong></p>
<p>📅 <strong>Schedule</strong>: At any time (no schedule defined).</p>
<p>🚦 <strong>Automerge</strong>: Disabled by config. Please merge this manually once you are satisfied.</p>
<p>♻️ <strong>Rebasing</strong>: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.</p>
<p>🔕 <strong>Ignore</strong>: Close this PR and you won't be reminded about this update again.</p>
<hr />
<p>☐ If you want to rebase/retry this PR, check this box.</p>
<hr />
<p>This PR has been generated by <a href="https://renovate.whitesourcesoftware.com">WhiteSource Renovate</a>. View repository job log <a href="https://app.renovatebot.com/dashboard#github/cucumber/common">here</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-16 15:40:12
[cucumber/common] Comment on #1688
<p>The current strategy to mark the commit to be released with a temporary <code>pre-release/**</code> branch is a bit fragile as, if that branch is deleted before the release is made, we don't know what commit to tag.</p>
<p>Maybe the pre-release script needs to make a "Draft" Release (and a tag), then we can just publish that Release when we make the actual release.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-16 15:44:14
[cucumber/common] Pull request merged by mattwynne
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-16 15:54:33
[cucumber/common] Comment on #1688
<p>We can pull out the sha of the commit being released like this:</p>
<pre><code>url=$(gh pr list --state closed --base release/v0.8.0 --json url | jq -r '.[0].url')
sha=$(gh pr view $url --json commits --jq '.commits[-1].oid')
</code></pre>
<p>That should work even if the <code>pre-release/**</code> branch has been deleted.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-17 09:49:07
[cucumber/common] Pull request opened by mpkorstanje
<p><strong>Summary</strong></p>
<p>Released <code>cucumber-expressions</code> from the wrong branch. Merging this branch should fix that.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-17 09:53:14
[cucumber/common] Pull request merged by mpkorstanje
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-18 01:20:24
[cucumber/common] Pull request opened by lee1990hayes
<p>Json</p>
<p><strong>Summary</strong>
<strong>Details</strong>
<strong>Motivation and Context</strong>
<strong>How Has This Been Tested?</strong>
<strong>Screenshots (if appropriate):</strong>
<strong>Types of changes</strong></p>
<p>☑︎ Bug fix (non-breaking change which fixes an issue).
☑︎ New feature (non-breaking change which adds functionality).
☑︎ Breaking change (fix or feature that would cause existing functionality to not work as expected).</p>
<p><strong>Checklist:</strong></p>
<p>☑︎ The change has been ported to Java.
☑︎ The change has been ported to Ruby.
☑︎ The change has been ported to JavaScript.
☑︎ The change has been ported to Go.
☑︎ The change has been ported to .NET.
☑︎ I've added tests for my code.
☑︎ My change requires a change to the documentation.
☑︎ I have updated the documentation accordingly.
☑︎ I have updated the CHANGELOG accordingly.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-18 01:21:41
[cucumber/common] Comment on #1701
<p>Json X vs Stig a</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-08-18 01:21:49
[cucumber/common] Pull request closed by lee1990hayes
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2021-10-21 13:53:17
[cucumber/vscode] is now public!
<a href="https://github.com/cucumber/vscode">cucumber/vscode</a>
ciaranmcnulty
2022-03-24 06:01:00
*Thread Reply:* Woohoo ```% composer require cucumber/messages Info from https://repo.packagist.org: #StandWithUkraine Using version ^18.0 for cucumber/messages ./composer.json has been updated Running composer update cucumber/messages Loading composer repositories with package information Info from https://repo.packagist.org: #StandWithUkraine Updating dependencies Lock file operations: 1 install, 0 updates, 0 removals
- Locking cucumber/messages (v18.0.0) Writing lock file Installing dependencies from lock file (including require-dev) Package operations: 1 install, 0 updates, 0 removals
- Downloading cucumber/messages (v18.0.0)
- Installing cucumber/messages (v18.0.0): Extracting archive Generating autoload files```
ciaranmcnulty
2022-03-24 07:07:48
*Thread Reply:* Looks ok. Only weirdness is php's updating the gherkin constraint on post-release vs release
GitHub
2022-11-03 06:00:24
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2115#issuecomment-1302064717">Comment</a> on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p>Unfortunately this looks like a feature to me:</p>
<p><a href="https://github.com/cucumber/common/blob/da2508f1edc4276933cba0f333a37026568772e4/gherkin/java/src/main/java/io/cucumber/gherkin/GherkinLine.java#L98-L114">common/gherkin/java/src/main/java/io/cucumber/gherkin/GherkinLine.java</a></p>
<p>Lines 98 to 114 in </cucumber/common/commit/da2508f1edc4276933cba0f333a37026568772e4|da2508f></p>
<p>Though since the escape is invalid, I think you might be okay mapping every <code>\</code> to <code>\\</code>. I don't immediately see any clashes.</p>
<p>So reformatting:</p>
<pre><code>| \o\no\ |
</code></pre>
<p>Should yield</p>
<pre><code>| \\o\no\\ |
</code></pre>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-11-03 07:14:55
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2115#issuecomment-1302179006">Comment</a> on issue by <a href="https://github.com/jdeniau">jdeniau</a>
<p>OK thank you for your fast response.</p>
<p>One think I find weird then (but that possibly because I do not know the implementation details here) is the fact that there is a difference between <code>\|</code> and the rest : <code>\|</code> are converted to <code>|</code>. But after reading my phrase several times, it seems logical as you want the cell value here.</p>
<p>About mapping <code>\</code> to <code>\\</code>, that's what I did, but I think that Behat does not escape the same way as cucumber handles it, so changing from <code>\</code> to <code>\\</code> in gherkin files does crash on the Behat side (I might open an issue about this when I do investigate a little further on their side).</p>
<p>For the moment I'm still stucked on the prettier plugin, and the hack I found is not pretty 😰</p>
<p><a href="https://github.com/mapado/prettier-plugin-gherkin/blob/main/src/GherkinAST/index.ts#L46-L59">https://github.com/mapado/prettier-plugin-gherkin/blob/main/src/GherkinAST/index.ts#L46-L59</a></p>
<p><a href="https://github.com/mapado/prettier-plugin-gherkin#escapebackslashes">https://github.com/mapado/prettier-plugin-gherkin#escapebackslashes</a></p>
<p>But it does work for the moment, so that's fine 👍</p>
<p>As an evolution, what do you think about adding the initial data in the AST ? or at least the a <code>endColumn</code> information in the <code>Location</code> of every nodes ? This way we might get the original data when we want (I can not actually do that because the size of the <code>TableCell</code> is lost as the <code>value</code> is transformed).</p>
<p>Thanks again</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-11-03 08:46:54
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2115#issuecomment-1302309968">Comment</a> on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p>> About mapping \ to \, that's what I did, but I think that Behat does not escape the same way as cucumber handles it, so changing from \ to \ in gherkin files does crash on the Behat side (I might open an issue about this when I do investigate a little further on their side).</p>
<p>I think that would be the way forward. I know that <a href="https://github.com/ciaranmcnulty">@ciaranmcnulty</a> contributes to both Cucumber Gherkin and <a href="https://github.com/Behat/Gherkin">Behat Gherkin</a>. Perhaps he can help out.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-11-04 03:55:09
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2115#issuecomment-1303256994">Comment</a> on issue by <a href="https://github.com/ciaranmcnulty">ciaranmcnulty</a>
<p>Behat currently uses its own parser which isn't 100% compatible with the Cucumber one. We do attempt to test against the same testdata but <a href="https://github.com/Behat/Gherkin/blob/master/tests/Behat/Gherkin/Cucumber/CompatibilityTest.php#L28">exclude some of the example features</a> . The data table examples don't seem to be in that exclusion list, which indicates the Behat parser produces the same result as Cucumber for those testdata.</p>
<p>Therefore a good step would be to add an example with this problematic case to the testdata in this repo and canonicalise the correct parsing. This would help us ensure that other implementations are parsing the source in the same way too.</p>
<p>The good news is that <a href="https://github.com/Behat/Behat/pull/1404">Behat should use the new parser at some point</a> which will mean the parsing will be much more compatible, and maintained in this organisation</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-11-07 06:47:49
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2115#issuecomment-1305723171">Comment</a> on issue by <a href="https://github.com/jdeniau">jdeniau</a>
<p><a href="https://github.com/ciaranmcnulty">@ciaranmcnulty</a> thank you for your response.<br />
The file I did work with is <code>escaped_pipes.feature</code> that is <a href="https://github.com/Behat/Gherkin/blob/master/tests/Behat/Gherkin/Cucumber/CompatibilityTest.php#L40">in your list</a>. I did not remember all other files.</p>
<p>I did add <a href="https://github.com/mapado/prettier-plugin-gherkin/blob/main/tests/plugin/unescape_backslashes_table_cell.feature">a file with lots of <code>\</code></a> in my test stack and the actual output of the prettier plugin <a href="https://github.com/mapado/prettier-plugin-gherkin/blob/main/tests/plugin/__snapshots__/jsfmt.spec.js.snap#L258-L304">is visible here</a> (but with backslashes escaped, which is nearly unreadable 😄 ).</p>
<p>I think the fix for the prettier plugin should work for now. I will watch carefully the fact that behat will use the cucumber parser and test it once released.</p>
<p>I'm closing this as <a href="https://github.com/mapado/prettier-plugin-gherkin#escapebackslashes">the option in the prettier plugin</a> should handle most cases.</p>
<p>Thanks <a href="https://github.com/mpkorstanje">@mpkorstanje</a> too for your time.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-11-04 12:48:05
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2116#issuecomment-1304092953">Comment</a> on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p>I believe that we already have some of this functionality for Java and Javascript. It doesn't exist (yet) for Python.</p>
<p>I reckon a pull requests would be welcome.</p>
<p><a href="https://github.com/cucumber/common/tree/main/gherkin-utils#prettygherkindocument-messagesgherkindocument-syntax-gherkin--markdown">https://github.com/cucumber/common/tree/main/gherkin-utils#prettygherkindocument-messagesgherkindocument-syntax-gherkin--markdown</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-11-08 05:11:08
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/1273#issuecomment-1307196036">Comment</a> on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p>No MCVE provided in two years. Please feel free to open a new issue with an MCVE.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-11-21 14:54:29
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2126#issuecomment-1322763950">Comment</a> on issue by <a href="https://github.com/mattwynne">mattwynne</a>
<p>Done: <a href="https://github.com/cucumber/query">https://github.com/cucumber/query</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-11-19 09:21:08
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2127#issuecomment-1320928604">Comment</a> on issue by <a href="https://github.com/mattwynne">mattwynne</a>
<p>Code and tags migrated: <a href="https://github.com/cucumber/gherkin-utils">https://github.com/cucumber/gherkin-utils</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-11-21 09:31:46
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2127#issuecomment-1322417395">Comment</a> on issue by <a href="https://github.com/mattwynne">mattwynne</a>
<p>Done!</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-11-10 04:08:07
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/871#issuecomment-1310188158">Comment</a> on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p>I've found documentation for the JUnit XML format:</p>
<p><a href="https://llg.cubic.org/docs/junit/">https://llg.cubic.org/docs/junit/</a><br />
<a href="https://github.com/junit-team/junit5/blob/43638eb6a870e0d6c49224053dfeb39dcf0ef33f/platform-tests/src/test/resources/jenkins-junit.xsd">https://github.com/junit-team/junit5/blob/43638eb6a870e0d6c49224053dfeb39dcf0ef33f/platform-tests/src/test/resources/jenkins-junit.xsd</a><br />
<a href="https://github.com/junit-team/junit5/issues/86">junit-team/junit5#86</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-11-21 15:29:33
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2134#issuecomment-1322795488">Comment</a> on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p>I can't find it in the docs right now but we can have a org level README on GitHub. It will show up prominently to people who visit the org.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-11-21 15:48:38
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2134#issuecomment-1322809350">Comment</a> on issue by <a href="https://github.com/davidjgoss">davidjgoss</a>
<p><a href="https://docs.github.com/en/organizations/collaborating-with-groups-in-organizations/customizing-your-organizations-profile#adding-a-public-organization-profile-readme">https://docs.github.com/en/organizations/collaborating-with-groups-in-organizations/customizing-your-organizations-profile#adding-a-public-organization-profile-readme</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-12-02 16:42:29
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2138#issuecomment-1335989559">Comment</a> on issue by <a href="https://github.com/mattwynne">mattwynne</a>
<p>Done: <a href="https://github.com/cucumber/action-publish-pypi">https://github.com/cucumber/action-publish-pypi</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-12-02 05:52:42
*Thread Reply:*
New comment on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p><code>This comment was deleted</code></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-12-02 06:00:36
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2139#issuecomment-1335270896">Comment</a> on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p>We can also limit the actions allowed in the organisation in a few different ways</p>
<p>• Only allow actions, cucumber org actions.
• Only allow actions, cucumber org actions and actions from verified publishers.</p>
<p>We currently have these non-cucumber non-github provided actions, I can't tell which ones are verified publishers:</p>
<pre><code>mpkorstanje@nyx:~/Projects/cucumber/code-search$ grep -r uses: | grep ".github" | cut -d ":" -f 3 | sort | uniq | grep -v cucumber | grep -v actions
8398a7/action-slack@v3
arduino/setup-protoc@v1
aurelien-baudet/workflow-dispatch@v2
codecov/codecov-action@v1
codecov/codecov-action@v3
coverallsapp/github-action@master
dart-lang/setup-dart@v1.3
docker/bake-action@v2
docker/build-push-action@v3
docker/login-action@v2
docker/setup-buildx-action@v2
docker/setup-qemu-action@v2
erlef/setup-beam@v1
GabrielBB/xvfb-action@v1
golangci/golangci-lint-action@v3.2.0
goreleaser/goreleaser-action@v2
goreleaser/goreleaser-action@v3.1.0
HaaLeo/publish-vscode-extension@v1
marocchino/sticky-pull-request-comment@v2
mymindstorm/setup-emsdk@v11
ocaml/setup-ocaml@v2
pulumi/setup-pulumi@v2
reactivecircus/android-emulator-runner@v2
ruby/setup-ruby@v1
shivammathur/setup-php@v2
snok/install-poetry@v1
softprops/action-gh-release@v1
</code></pre>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-12-02 06:15:28
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2139#issuecomment-1335296519">Comment</a> on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p>Projects that use the <code>cucumber/action-create-github-release</code> that would definitely need elevated permissions.</p>
<pre><code>mpkorstanje@nyx:~/Projects/cucumber/code-search$ grep -rl cucumber/action-create-github-release | cut -d '/' -f 1
cucumber-expressions
blockly
cucumber-parent
action-get-versions
gherkin
message-streams
action-publish-rubygem
action-publish-sbt
action-publish-nuget
action-publish-hex
action-create-github-release
action-create-github-release
action-create-github-release
action-create-github-release
action-create-github-release
action-create-github-release
gherkin-streams
cucumber-jvm-scala
action-publish-npm
multi_test
html-formatter
compatibility-kit
action-create-release-pr
cucumber-js-pretty-formatter
ci-environment
cucumber-js
react-components
language-server
action-publish-subrepo
screenplay.js
build
cucumber-ruby
language-service
cucumber-ruby-wire
cucumber-json-converter
monaco
cucumber-android
messages
action-publish-mvn
gherkin-utils
action-publish-pypi
microdata
split-java
tag-expressions
query
action-publish-cpan
cucumber-ruby-core
cucumber-rails
release-tests
cucumber-jvm
</code></pre>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-12-08 00:02:16
*Thread Reply:*
<a href="https://github.com/cucumber/common/pull/1013#issuecomment-1342232880">Comment</a> on pull request by <a href="https://github.com/apps/renovate">renovate[bot]</a>
<p><strong>Renovate is disabled</strong></p>
<p>Renovate is disabled due to lack of config. If you wish to reenable it, you can either (a) commit a config file to your base branch, or (b) rename this closed PR to trigger a replacement onboarding PR.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-01-10 10:50:03
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2140#issuecomment-1377701876">Comment</a> on issue by <a href="https://github.com/davidjgoss">davidjgoss</a>
<p>Happy to take a look at this for cucumber-js at least. I personally like having a bot to close/lock issues, it irks people less than a human doing it.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-01-12 07:43:05
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2140#issuecomment-1380585382">Comment</a> on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p>I'll delete the files from all the projects. Then we can set it up again where we want it.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-12-27 12:56:20
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/871#issuecomment-1366176873">Comment</a> on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p>Java is done. Next step would be to extract the formatter added in <a href="https://github.com/cucumber/cucumber-js/pull/2121">cucumber/cucumber-js#2121</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-12-27 14:10:17
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2141#issuecomment-1366205377">Comment</a> on issue by <a href="https://github.com/davidjgoss">davidjgoss</a>
<p>FWIW cucumber-js will release a major version in January that drops node 12 support, at that point all the libraries can safely follow suit.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2022-12-28 02:25:08
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2141#issuecomment-1366544125">Comment</a> on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p>Cheers!</p>
<p>14 is also going to drop off support soonish.</p>
<p>And to keep the size of the CI matrix under control I think it would also be sensible to only test against the LTS versions (so 16, 18).</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-01-03 13:46:03
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2142#issuecomment-1370255952">Comment</a> on issue by <a href="https://github.com/mattwynne">mattwynne</a>
<p>Thanks <a href="https://github.com/mpkorstanje">@mpkorstanje</a>! I'll get on with removing Dependabot everywhere.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-01-03 15:29:55
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2143#issuecomment-1370328442">Comment</a> on issue by <a href="https://github.com/mattwynne">mattwynne</a>
<p><a href="https://github.com/cucumber/aruba/pull/886">cucumber/aruba#886</a> is failing, I'm not going to wait on it.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-01-16 01:30:52
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2145#issuecomment-1383741879">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p>This is a question for software that exists outside of cucumber. Your best bet is to look into using a tool to do this. In java - the cluecumber plugin: <a href="https://github.com/trivago/cluecumber">https://github.com/trivago/cluecumber</a> might be your best starting point</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-02-09 09:30:03
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/1063#issuecomment-1424558115">Comment</a> on issue by <a href="https://github.com/rohalravi">rohalravi</a>
<p>Can anyone please help me, I am also getting the same issue , I am using these gems<br />
gem 'gherkin'<br />
gem 'cucumber', '~> 3.1.2'<br />
gem 'cuke_modeler', '~> 3.3.0'</p>
<p>Error:<br />
wrong number of arguments (given 0, expected 1) (ArgumentError)<br />
C:/Ruby25-x64/lib/ruby/gems/2.5.0/gems/cucumber-gherkin-13.0.0/lib/gherkin/ast_builder.rb:6:in <code>initialize' C:/Ruby25-x64/lib/ruby/gems/2.5.0/gems/cucumber-gherkin-13.0.0/lib/gherkin/parser.rb:59:in</code> new'<br />
C:/Ruby25-x64/lib/ruby/gems/2.5.0/gems/cucumber-gherkin-13.0.0/lib/gherkin/parser.rb:59:in <code>initialize' C:/Ruby25-x64/lib/ruby/gems/2.5.0/gems/cucumber-core-3.2.1/lib/cucumber/core/gherkin/parser.rb:23:in</code> new'<br />
C:/Ruby25-x64/lib/ruby/gems/2.5.0/gems/cucumber-core-3.2.1/lib/cucumber/core/gherkin/parser.rb:23:in <code>document' C:/Ruby25-x64/lib/ruby/gems/2.5.0/gems/cucumber-core-3.2.1/lib/cucumber/core.rb:30:in</code> block in parse'<br />
C:/Ruby25-x64/lib/ruby/gems/2.5.0/gems/cucumber-core-3.2.1/lib/cucumber/core.rb:29:in <code>each' C:/Ruby25-x64/lib/ruby/gems/2.5.0/gems/cucumber-core-3.2.1/lib/cucumber/core.rb:29:in</code> parse'<br />
C:/Ruby25-x64/lib/ruby/gems/2.5.0/gems/cucumber-core-3.2.1/lib/cucumber/core.rb:21:in <code>compile' C:/Ruby25-x64/lib/ruby/gems/2.5.0/gems/cucumber-3.1.2/lib/cucumber/runtime.rb:75:in</code> run!'<br />
C:/Ruby25-x64/lib/ruby/gems/2.5.0/gems/cucumber-3.1.2/lib/cucumber/cli/main.rb:34:in <code>execute!' C:/Ruby25-x64/lib/ruby/gems/2.5.0/gems/cucumber-3.1.2/bin/cucumber:9:in</code> <top (required)>'<br />
C:/Ruby25-x64/bin/cucumber:23:in <code>load' C:/Ruby25-x64/bin/cucumber:23:in</code></p>
<p>'</p>
<p>I tried</p>
<p><strong>Keep the default argument</strong></p>
<p>def initialize(ast_builder = AstBuilder.new(Cucumber::Messages::IdGenerator::UUID.new))</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-02-09 09:55:21
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/1063#issuecomment-1424591478">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p>You are likely using the conflicting gems for gherkin. Ideally ensure you don't install gherkin manually (Although it is possible).</p>
<p>There is a <code>cucumber-gherkin</code> gem which has replaced <code>gherkin</code></p>
<p>As you are using an old version of cucumber (v3), I'd also advise you to upgrade that if possible. As we don't support legacy versions of the software currently.</p>
<p>If you have an issue still after this, please open a new issue</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-02-23 17:42:31
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2147#issuecomment-1442676044">Comment</a> on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p>I don't see Cucumber in that stacktrace. So probably a problem with Eclipse. Did you see this problem on Stackoverflow?</p>
<p><a href="https://stackoverflow.com/questions/33521436/junit-noclassdeffounderror-org-junit-runner-manipulation-filter">https://stackoverflow.com/questions/33521436/junit-noclassdeffounderror-org-junit-runner-manipulation-filter</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-02-23 17:47:20
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2147#issuecomment-1442679202">Comment</a> on issue by <a href="https://github.com/ank1006">ank1006</a>
<p>yep, saw that but surprisingly i don't see ModulePath and Classpath within libraries tab when i navigate to Java Build Path. Within libraries i directly see dependencies i added, JRE System Library and the Junit 4.<br />
I just updated package names, removed entry from run/debug setting of my runner file, restarted eclipse and now on running i am seeing following exception-</p>
<p>java.lang.NullPointerException<br />
at java.time.Duration.between(Unknown Source)<br />
at io.cucumber.core.runtime.CucumberExecutionContext.emitTestRunFinished(CucumberExecutionContext.java:129)<br />
at io.cucumber.core.runtime.CucumberExecutionContext.finishTestRun(CucumberExecutionContext.java:118)<br />
at io.cucumber.junit.Cucumber$FinishTestRun.evaluate(Cucumber.java:250)<br />
at org.junit.runners.ParentRunner$3.evaluate(ParentRunner.java:306)<br />
at org.junit.runners.ParentRunner.run(ParentRunner.java:413)<br />
at org.eclipse.jdt.internal.junit4.runner.JUnit4TestReference.run(JUnit4TestReference.java:89)<br />
at org.eclipse.jdt.internal.junit.runner.TestExecution.run(TestExecution.java:41)<br />
at org.eclipse.jdt.internal.junit.runner.RemoteTestRunner.runTests(RemoteTestRunner.java:541)<br />
at org.eclipse.jdt.internal.junit.runner.RemoteTestRunner.runTests(RemoteTestRunner.java:763)<br />
at org.eclipse.jdt.internal.junit.runner.RemoteTestRunner.run(RemoteTestRunner.java:463)<br />
at org.eclipse.jdt.internal.junit.runner.RemoteTestRunner.main(RemoteTestRunner.java:209)</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-02-28 04:53:52
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2141#issuecomment-1448129415">Comment</a> on issue by <a href="https://github.com/davidjgoss">davidjgoss</a>
<p>cucumber.js 9.0.0 is out and so our libraries can freely remove Node.js 12 support now. I'm happy to pick that up.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-03-25 16:43:18
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/1410#issuecomment-1483948380">Comment</a> on issue by <a href="https://github.com/psrini7">psrini7</a>
<p><a href="https://github.com/pradipyogipacific">@pradipyogipacific</a> I’m facing similar issue could you please let me know how you fixed it ?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-03-27 09:42:48
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2148#issuecomment-1485470469">Comment</a> on issue by <a href="https://github.com/davidjgoss">davidjgoss</a>
<p>Thanks for reporting - I’ve flagged this up to the admins of our Slack and will post further updates here.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-03-31 08:05:04
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2148#issuecomment-1492077797">Comment</a> on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p>Matt provided a work around for this.</p>
<p>Integrated <a href="https://protect-us.mimecast.com/s/xB5JCmZjk8CAvYRvTGKlV1?domain=communityinviter.com">https://communityinviter.com</a> into our Slack and fixed the link on <a href="https://cucumber.io/docs/community/get-in-touch/">https://cucumber.io/docs/community/get-in-touch/</a> to point to it.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-04-17 06:36:38
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2149#issuecomment-1511363576">Comment</a> on issue by <a href="https://github.com/davidjgoss">davidjgoss</a>
<p>> if the Scenario Outline would accept arguemts as the When statement does</p>
<p>Have you tried this? I'm pretty certain it works like this already. We might just need an update to the docs here.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-04-18 01:32:24
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2149#issuecomment-1512678968">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p>One thing to bear in mind is this sometimes can trip up on reporters. But this functionality 100% exists.</p>
<p>I personally don't advocate this, but there's nothing stopping you doing it. I would use something like</p>
<p>> Scenario Outline: Successfully add a variety of user-types</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-04-18 03:16:22
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2149#issuecomment-1512818298">Comment</a> on issue by <a href="https://github.com/BeyerMart">BeyerMart</a>
<p>Great. Thank you.<br />
I didn't manage to get the argument working in the Scenario Outline, but follwoing your input, I will rename my test scenario to be more generic.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-05-16 10:04:45
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2150#issuecomment-1550050711">Comment</a> on issue by <a href="https://github.com/saipavanmeruga">saipavanmeruga</a>
<p>I can help you in updating the yaml and script files. Can you assign me this issue?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-06-28 09:13:38
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2151#issuecomment-1611720350">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p>This looks like you're using some form of additional plugin / framework (Possibly a cypress preprocessor, I'm not sure).</p>
<p><code>Feature</code> as a keyword is designed to appear once at the top of each feature file <code>**.feature</code>, to indicate what feature it is you are describing (Possibly testing). It's design is a little more than just grouping scenarios, as other keywords (Such as Rule), can do this behaviour also.</p>
<p>Ideally you would use multiple feature files for multiple features, thus keeping your features separate. If you need to describe nuances of your features, we have a keyword <code>Rule</code> that acts as a wrapper to partition up your Feature</p>
<p>If you would want to suggest changes to the gherkin language, such as this, then the gherkin repo would be best. However advanced warning, something like this is quite a large change, and would need to satisfy a large subsection of the gherkin-using community. Please bear that in mind.</p>
<p>The gherkin repo is here: <a href="https://github.com/cucumber/gherkin">https://github.com/cucumber/gherkin</a></p>
<p>I'll leave this open for now to await your reply</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-06-28 09:47:27
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2151#issuecomment-1611763964">Comment</a> on issue by <a href="https://github.com/jasp402">jasp402</a>
<p>Indeed, I'm using Cypress in conjunction with a preprocessor, specifically the <a href="https://github.com/badeball/cypress-cucumber-preprocessor">cypress-cucumber-preprocessor</a>. I've just made this request in the corresponding repository. However, it would be interesting to delve into how the same result could be achieved using rules. It's also worth noting that this might not be a valid solution for me if the preprocessor has limitations.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-06-28 11:45:17
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2151#issuecomment-1611908649">Comment</a> on issue by <a href="https://github.com/jkronegg">jkronegg</a>
<p>I agree with <a href="https://github.com/luke-hill">@luke-hill</a>: this looks like a need that can be solved using a plugin.</p>
<p>I have a similar need of making reports which groups several Features into sections (without cypress). I made a plugin wich groups Feature based on their classpath directory and custom annotations in the feature files.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-06-28 12:57:33
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2151#issuecomment-1612020599">Comment</a> on issue by <a href="https://github.com/jenisys">jenisys</a>
<p><a href="https://github.com/jasp402">@jasp402</a><br />
Have you considered the <code>Rule</code> keyword for grouping scenarios ?</p>
<p>NOTE: That would allow you one additional level of grouping below Feature.</p>
<p>SEE ALSO:</p>
<p>• <a href="https://docs.cucumber.io/docs/gherkin/reference/">https://docs.cucumber.io/docs/gherkin/reference/</a> (Section: Rule)</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-06-28 13:42:15
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2151#issuecomment-1612078328">Comment</a> on issue by <a href="https://github.com/systemsincode">systemsincode</a>
<p>Have you tried using <a href="https://github.com/tagging">@tagging</a> if you need to selectively run subset of feature file scenarios for example and helps provide meta info to help describe scope of tests.</p>
<p><#|…></p>
<p>On Wed, 28 Jun 2023, 20:57 jenisys, <strong><em>*</strong>@<strong></em><em></strong>.<strong></em><em></strong>> wrote: <a href="https://github.com/jasp402">@jasp402</a> <<a href="https://github.com/jasp402">https://github.com/jasp402</a>> Have you considered the Rule keyword for grouping scenarios ? NOTE: That would allow you one additional level of grouping below Feature. SEE ALSO: - <a href="https://docs.cucumber.io/docs/gherkin/reference/">https://docs.cucumber.io/docs/gherkin/reference/</a> (Section: Rule) — Reply to this email directly, view it on GitHub <<a href="https://github.com/cucumber/common/issues/2151#issuecomment-1612020599">#2151 (comment)</a>>, or unsubscribe <<a href="https://github.com/notifications/unsubscribe-auth/ADLFFUEERPBDPANWNSBY7LTXNSEEBANCNFSM6AAAAAAZXKZKZI">https://github.com/notifications/unsubscribe-auth/ADLFFUEERPBDPANWNSBY7LTXNSEEBANCNFSM6AAAAAAZXKZKZI</a>> . You are receiving this because you are subscribed to this thread.Message ID: <strong></em><em></strong>@<strong></em><em></strong>.<strong></em>*</strong>></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-07-09 02:54:51
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2152#issuecomment-1627664890">Comment</a> on issue by <a href="https://github.com/jenisys">jenisys</a>
<p>Mmh, your expectations are at least partly wrong (IMHO):</p>
<p>• If a step fails in a Scenario, the remaining steps of this Scenario are normally <code>skipped</code> (but: the <code>scenario.status</code> is still <code>failed</code>). Therefore, the <code>step.status = skipped</code> reflects exactly what occurs internally
• There are some cases (or: <code>cucumber</code> implementations) that can be configured to continue after the first <code>failed</code> step. In these cases, the <code>step.status</code> of the remaining steps after the first <code>failed</code> step are not <code>skipped</code> but are <code>passed</code> or <code>failed</code> whatever occurs when executing these remaining steps.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-07-09 03:05:23
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2152#issuecomment-1627667115">Comment</a> on issue by <a href="https://github.com/richardforjoesky">richardforjoesky</a>
<p>Thanks <a href="https://github.com/jenisys">@jenisys</a> , was hoping to get some of my expectations clarified, as i wasn't 100% sure myself. Regarding skipped. Since the steps are being skipped because the step / scenario failed. Feels like it should be failed, since the reason for it being skipped is because the step failed?</p>
<p>So<br />
Scenario</p>
<p>• step - passed
• step - failed
• step skipped<br />
=> scenario failed</p>
<p>Scenario</p>
<p>• step - skipped
• step - skipped
• step - skipped<br />
=> scenario skipped</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-07-09 03:34:04
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2152#issuecomment-1627674484">Comment</a> on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p>Generally speaking the way to determine the result of a scenario is to take the worst status of its steps.</p>
<p><a href="https://github.com/cucumber/cucumber-jvm/blob/main/cucumber-core/src/main/java/io/cucumber/core/runner/TestCaseState.java#L54C1-L54C1">https://github.com/cucumber/cucumber-jvm/blob/main/cucumber-core/src/main/java/io/cucumber/core/runner/TestCaseState.java#L54C1-L54C1</a></p>
<p>The order being</p>
<pre><code> PASSED,
SKIPPED,
PENDING,
UNDEFINED,
AMBIGUOUS,
FAILED
</code></pre>
<p><a href="https://github.com/cucumber/cucumber-jvm/blob/main/cucumber-java/src/main/java/io/cucumber/java/Status.java">https://github.com/cucumber/cucumber-jvm/blob/main/cucumber-java/src/main/java/io/cucumber/java/Status.java</a></p>
<p>If a scenario has no steps implementations currently vary a bit on what the outcome should be.</p>
<p><a href="https://github.com/cucumber/gherkin/issues/11">cucumber/gherkin#11</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-07-09 03:34:47
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2152#issuecomment-1627674757">Comment</a> on issue by <a href="https://github.com/jenisys">jenisys</a>
<p>HINT (just that we both have the correct understanding; as <a href="https://github.com/mpkorstanje">@mpkorstanje</a> stated):<br />
The step <code>skipped</code> does not cause the <code>scenario.status = failed</code> but any <code>step.status = failed</code> does (caused by the <code>failing step</code> that occurred first).<br />
The <code>step.status = skipped</code> is just a correlated change caused by the <code>failing step</code>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-07-09 03:50:34
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2152#issuecomment-1627679086">Comment</a> on issue by <a href="https://github.com/richardforjoesky">richardforjoesky</a>
<p>Yea that sounds right to me. failed is caused by worst status</p>
<p>So even though the last step is skipped, since the worst status is failed from previous steps, final status should be failed, rather than skipped. based on <a href="https://github.com/mpkorstanje">@mpkorstanje</a> ordering</p>
<p>Scenario<br />
step - passed<br />
step - failed<br />
step skipped<br />
=> scenario failed (worst status)</p>
<p>Scenario<br />
step - passed<br />
step - passed<br />
step - skipped<br />
=> scenario skipped (worst status)</p>
<p>Scenario<br />
step - skipped<br />
step - passed<br />
step - pending<br />
=> scenario pending (worst status)</p>
<p>Scenario<br />
step - passed<br />
step - undefined<br />
step - passed<br />
=> scenario undefined (worst status)</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-07-09 05:30:09
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2152#issuecomment-1627702240">Comment</a> on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p>Since you're using <code>@badeball/cypress-cucumber-preprocessor</code> you'll want to take this issue there.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-07-09 05:43:23
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2152#issuecomment-1627705432">Comment</a> on issue by <a href="https://github.com/richardforjoesky">richardforjoesky</a>
<p>Ooops, sorry about that. Accidentally raised it here. Thanks for your feedback</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-08-10 12:08:25
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2153#issuecomment-1673762269">Comment</a> on issue by <a href="https://github.com/davidjgoss">davidjgoss</a>
<p>Could you link to the page you’re seeing the issue on please <a href="https://github.com/nirupagajera">@nirupagajera</a>?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-08-11 00:30:52
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/960#issuecomment-1674322108">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p>Done above</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-08-11 03:13:02
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2150#issuecomment-1674517813">Comment</a> on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p>Hey thanks for offering to help. Unfortunately all the <code>action-publish-**</code> repos can only be changed by core members. This makes it a less then stellar first issue.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-08-21 12:59:26
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/568#issuecomment-1686957830">Comment</a> on issue by <a href="https://github.com/pgundlupetvenkatesh">pgundlupetvenkatesh</a>
<p>Been using Cucumber-ruby extensively at <a href="https://www.exlservice.com/industries/health">EXL Service Health</a> from 2017 mainly for Web UI testing with Watir.</p>
<p><a href="https://user-images.githubusercontent.com/9060416/262139652-24e3845d-70ff-4d3e-8fcc-6342750900e9.jpg">exl_logo</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-08-30 02:04:37
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2154#issuecomment-1698781323">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p>This looks similar to steps in steps in ruby, which is something we're advising heavily against.</p>
<p>Ideally if you have lots of common steps, you should use that languages abstraction techniques. So in ruby this would be a module or a class, and similar in Java.</p>
<p>Then you would call the methods in the steps. so something like.</p>
<pre><code>Given('I log in as user {word} with password {word}') do |user, password|
# method here to login, using a class or module
end
</code></pre>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-08-30 02:09:50
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2154#issuecomment-1698789597">Comment</a> on issue by <a href="https://github.com/ehuelsmann">ehuelsmann</a>
<p>This Stack Overflow thread: <a href="https://stackoverflow.com/a/23092261">https://stackoverflow.com/a/23092261</a></p>
<p>Refers to this discussion: <a href="https://groups.google.com/g/cukes/c/DzE_kGZx94I/m/5rf__N31qvAJ">https://groups.google.com/g/cukes/c/DzE<em>kGZx94I/m/5rf</em>_N31qvAJ</a></p>
<p>which can be summarized as "Gherkin is a collaboration tool, not a programming language". What makes your request different?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-08-31 18:37:14
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2154#issuecomment-1701991165">Comment</a> on issue by <a href="https://github.com/shihai1991">shihai1991</a>
<p>> Ideally if you have lots of common steps, you should use that languages abstraction techniques. So in ruby this would be a module or a class, and similar in Java.</p>
<p>Thanks for your suggestion. You are right. Using some computer languages' techniques can resolve this problem too. But I would prefer to use <code>StepMacros</code> to combine those steps. Some of my reasons:</p>
<ol><li>We have kinds of teams. Some developers use Java, others use Golang or C/C++. In order to make sure all of our teammates to understand the test scenario efficiently, we must discuss the scenario details in Gherkin language level. Involving kinds of computer languages' techniques will destroy the effectiveness;</li><li>Our test scenario is Cloud Computing. There are many Resource concepts. We will get a very long feature file If we describe all details in one feature. For example:</li>
</ol>
<pre><code># create VPC1
# Of Course, we can combine the multi step description in one step
# something like: When create a VPC with the the Availability region with the Availiability zone with ipv4...
# But the test scenario should be flexible and combinable. But the combined step lost the flexibility.
When build a VPC_Request_Body
And choose the Availability region in VPC_Request_Body
And choose the Availiability zone in VPC_Request_Body
And set the ipv4 in VPC_Request_Body
And set the subnet in VPC_Request_Body
And create an VPC with VPC_Request_Body
Then VPC created successful
# create VPC2
# the creation operation will be repeated in here, but the property ov VPC2 will be different
# create ECS1 in VPC1
# create ECS2 in VPC2
# test the connectivity between ECS1 and ECS2
</code></pre>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-08-31 19:58:05
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2154#issuecomment-1702071916">Comment</a> on issue by <a href="https://github.com/shihai1991">shihai1991</a>
<p>> Refers to this discussion: <a href="https://groups.google.com/g/cukes/c/DzE_kGZx94I/m/5rf__N31qvAJ">https://groups.google.com/g/cukes/c/DzE<em>kGZx94I/m/5rf</em>_N31qvAJ</a><br />
> which can be summarized as "Gherkin is a collaboration tool, not a programming language". What makes your request different?</p>
<p>Oh, thanks for your information. Looks like my request is not different with the discussion.</p>
<p>Some points from <a href="http://blog.josephwilk.net/ruby/cucumber-waves-goodbye-to-givenscenario.html">cucumber-waves-goodbye-to-givenscenario</a></p>
<p>> Difficult to understand isolated scenarios</p>
<p>I think we can't consider this problem If the IDE support to find and link the related scenario.<br />
The IDE is much better now than 10 years ago.</p>
<p>> Creating setups which pretend to be scenarios</p>
<p>If the user need use this feature. IMHO, the key word <code>GivenScenario</code> could be updated to <code>CombinedGiven</code>、<code>CombinedWhen</code> and so on.</p>
<p>> Noise in the output</p>
<p>We don't focus on the output until we meet the failed testcase.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-09-05 01:24:49
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2154#issuecomment-1706166044">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p>I'm going to close this. We've briefly spoken about it in the committers meeting and async, and given this is something we are strongly advising against - all languages have their own paradigms. We advise using those.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-09-05 01:27:17
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2144#issuecomment-1706169759">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p>Marking this as completed. The changelog is updated. The polyglot release tool works well, the releasing doc is now updated in MAIN, more releases are coming soon in patch updates to tidy up some code. I'm back in.</p>
<p>Next step/s would be to do the same on alternative repos such as rails, but I don't anticipate any issues</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-09-05 12:48:32
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2144#issuecomment-1707219710">Comment</a> on issue by <a href="https://github.com/mattwynne">mattwynne</a>
<p>Sweet, thanks for your work getting this rolling Luke</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-09-05 01:30:02
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/1798#issuecomment-1706174045">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p><a href="https://github.com/mattwynne">@mattwynne</a> any updates here?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-09-05 12:48:02
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/1798#issuecomment-1707219120">Comment</a> on issue by <a href="https://github.com/mattwynne">mattwynne</a>
<p>I haven't found any technology that can do it. I found <a href="https://github.com/repository-settings/app">https://github.com/repository-settings/app</a> but last time I looked it had a bunch of usability problems (e.g. <a href="https://github.com/repository-settings/app#security-implications">https://github.com/repository-settings/app#security-implications</a>) and wasn't under active development. It looks like things may have picked up again, so it could still be an option.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-09-05 01:31:22
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/1521#issuecomment-1706176069">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p>A fair amount of work has been done here recently. Where do we think we're at <a href="https://github.com/ehuelsmann">@ehuelsmann</a> / <a href="https://github.com/mpkorstanje">@mpkorstanje</a> / <a href="https://github.com/davidjgoss">@davidjgoss</a> <a href="https://github.com/jenisys">@jenisys</a> ?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-09-05 01:31:59
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/1426#issuecomment-1706176970">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p><a href="https://github.com/mattwynne">@mattwynne</a> going to close this one off given we're going to be dealing with so much other stuff this isn't something we need to tackle in the short to medium term</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-09-05 01:32:27
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/999#issuecomment-1706177581">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p><a href="https://github.com/davidjgoss">@davidjgoss</a> guessing this is a js specific issue?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-09-05 01:33:05
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/933#issuecomment-1706178543">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p>Individual languages can now implement this how they want. Ruby is fine and I think other languages are good too. Suggest this gets re-created if further issues are needed to be amended.</p>
<p>cc/ <a href="https://github.com/charlierudolph">@charlierudolph</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-09-05 04:09:03
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2141#issuecomment-1706414040">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p><a href="https://github.com/davidjgoss">@davidjgoss</a> are we happy to close this now? Probably should live in cucumber-js right?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-09-05 04:14:26
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/1116#issuecomment-1706421474">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p>Do we think this can be marked as closed now then <a href="https://github.com/mpkorstanje">@mpkorstanje</a> / <a href="https://github.com/davidjgoss">@davidjgoss</a></p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-09-05 04:17:43
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/1116#issuecomment-1706425814">Comment</a> on issue by <a href="https://github.com/davidjgoss">davidjgoss</a>
<p>I would say it lives in cucumber-ruby, given jvm doesn't have this issue and js has solved it in its own way. The original query was about how it works in other impls which is answered, so now it's a case of what - if anything - should be changed in ruby land.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-09-05 04:15:10
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/897#issuecomment-1706422346">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p><a href="https://github.com/davidjgoss">@davidjgoss</a> from the OP did we ever get around to renaming the 2 key methods for JS?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-09-05 04:18:49
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/897#issuecomment-1706427317">Comment</a> on issue by <a href="https://github.com/davidjgoss">davidjgoss</a>
<p>JS methods are named <code>attach</code> and <code>log</code> respectively - so yeah I think we're done here.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-10-08 03:55:58
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2150#issuecomment-1751997223">Comment</a> on issue by <a href="https://github.com/roskee">roskee</a>
<p><a href="https://github.com/mpkorstanje">@mpkorstanje</a> I can take this issue.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-10-16 22:29:48
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2155#issuecomment-1765695338">Comment</a> on issue by <a href="https://github.com/mattwynne">mattwynne</a>
<p>Sorry, this isn't currently supported. It would require pretty major changes to Gherkin to be able to read examples from an excel file.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2023-10-16 13:25:21
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/999#issuecomment-1765215387">Comment</a> on issue by <a href="https://github.com/davidjgoss">davidjgoss</a>
<p>We have no devDependencies of json-schema in the packages I've just looked through, so I think this issue has come out in the wash.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2024-01-08 04:49:11
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2148#issuecomment-1880945741">Comment</a> on issue by <a href="https://github.com/arietis">arietis</a>
<p>FYI, there are still some pages using the old link, it took me some time to find the new one.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2024-01-08 05:05:42
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2148#issuecomment-1880971177">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p><a href="https://github.com/arietis">@arietis</a> if you spot any links that are incorrect please feel free to edit them as a PR. You should be able to do this on the docs site by clicking at the bottom.</p>
<p>For anything on Github you should be able to click edit on the md file</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2024-01-08 05:07:24
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/1138#issuecomment-1880973865">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p><a href="https://github.com/davidjgoss">@davidjgoss</a> should this be closed and relevant issue(s), raised on the constituent repos if needed?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2024-01-08 05:09:19
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/1519#issuecomment-1880977537">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p><a href="https://github.com/gasparnagy">@gasparnagy</a> having played around a bit recently with this. Do we think this is still an issue, and if so should we move it to the messages repo?</p>
<p>I noticed things like location often are half filled (So not all the props are passed in), and things seem to be working well.</p>
<p>Do we think we need additional work here>?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2024-01-08 07:13:57
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/1519#issuecomment-1881202932">Comment</a> on issue by <a href="https://github.com/gasparnagy">gasparnagy</a>
<p>I haven't checked the JSON schema recently, so I don't know. All I wanted is that if we make a specification about the messages (like a JSON schema) that should be valid for all the JSON messages we send across. So if we don't typically send e.g. <code>description</code> than that field should not be marked as <code>required</code>.<br />
Aslak and Matt sees the JSON schema more as a conceptual description of the message structure, primarily for code, where the "required" makes sense. I see that more as a specification of the pure JSON data we send and hence the disagreement. But it's not super important for me, so we can leave it as it is and close this issue.<br />
I'm not aware of any tools that would actually do schema validation, so there is no concrete error anyway.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2024-01-08 05:09:51
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2140#issuecomment-1880978290">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p><a href="https://github.com/mpkorstanje">@mpkorstanje</a> is this completed?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2024-01-09 11:36:34
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2140#issuecomment-1883663743">Comment</a> on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p>Never got around to it.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2024-01-16 07:15:53
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2157#issuecomment-1893951917">Comment</a> on issue by <a href="https://github.com/binhtran432k">binhtran432k</a>
<p>Does an implementation of a Prettier plugin for Gherkin already exist, such as the one found at <a href="https://github.com/mapado/prettier-plugin-gherkin">https://github.com/mapado/prettier-plugin-gherkin</a>?</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2024-01-16 07:44:19
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2157#issuecomment-1894008470">Comment</a> on issue by <a href="https://github.com/luke-hill">luke-hill</a>
<p>Some of these plugins are very subjective. Each to their own - and each language will likely "bleed over" some of their thoughts to their language equivalent feature files.</p>
<p>I would encourage you to use whichever one/s make most sense to you, there are a few out there as linked.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2024-01-22 08:28:52
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2159#issuecomment-1904366589">Comment</a> on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p>I think you're in the wrong project. You probably meant <a href="https://github.com/gkushang/cucumber-html-reporter">https://github.com/gkushang/cucumber-html-reporter</a>.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2024-01-31 00:15:20
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2160#issuecomment-1918598326">Comment</a> on issue by <a href="https://github.com/davidjgoss">davidjgoss</a>
<p>Other maintainers might feel differently, but I don’t think is something we would add to Cucumber. You an always extend/wrap DataTable to do what you need though.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
GitHub
2024-01-31 01:54:46
*Thread Reply:*
<a href="https://github.com/cucumber/common/issues/2160#issuecomment-1918761887">Comment</a> on issue by <a href="https://github.com/mpkorstanje">mpkorstanje</a>
<p>There are <a href="https://github.com/cucumber/cucumber-jvm/tree/main/datatable">many ways a data table could be mapped to an object</a> , but using parameter types would be a conceptual mismatch.</p>
<p>Anyway, I think it would be more useful to start a feature request from a concrete real world example. I.e. the problem you're trying to solve, not the solution you imagine.</p>
<a href="https://github.com/cucumber/common">cucumber/common</a>
